[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-huggingface-dell-ai":3,"mdc-d884f6-key":35,"related-repo-huggingface-dell-ai":2824,"related-org-huggingface-dell-ai":2833},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":24,"repoUrl":25,"updatedAt":26,"license":27,"forks":28,"topics":29,"repo":30,"sourceUrl":33,"mdContent":34},"dell-ai","discover and deploy models on Dell hardware","Skill for interacting with Dell Enterprise Hub (DEH) - discovering models, exploring hardware platforms, generating deployment snippets, etc.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},"huggingface","Hugging Face","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fhuggingface.png",[12,15,18,21],{"name":9,"slug":13,"type":14},"hugging-face","tag",{"name":16,"slug":17,"type":14},"Hardware","hardware",{"name":19,"slug":20,"type":14},"Deployment","deployment",{"name":22,"slug":23,"type":14},"AI Infrastructure","ai-infrastructure",10,"https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fdell-ai","2026-07-24T05:42:03.032879",null,7,[],{"repoUrl":25,"stars":24,"forks":28,"topics":31,"description":32},[],"The official Python SDK and CLI for the Dell Enterprise Hub","https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fdell-ai\u002Ftree\u002FHEAD\u002Fskills\u002Fdell-ai","---\nname: dell-ai\ndescription: Skill for interacting with Dell Enterprise Hub (DEH) - discovering models, exploring hardware platforms, generating deployment snippets, etc.\n---\n\n# Dell AI\n\ndell-ai interacts with the Dell Enterprise Hub (DEH) — discovering models, exploring hardware platforms, generating deployment snippets, managing app catalog entries, checking system compatibility, etc.\n\n## When to use\n\n- User wants to find AI models available on Dell Enterprise Hub\n- User wants to find AI models available on DEH for their hardware platform\n- User needs to check which Dell hardware platforms support a given model\n- User wants to generate Docker or Kubernetes deployment commands for a model and then run it.\n- User wants a goodput-optimized deployment snippet (balanced, long-context, high-concurrency, performance).\n- User wants to **deploy** a model or app directly onto the local node (not just generate a snippet)\n- User wants to **tear down** \u002F undeploy a running deployment\n- User wants to **check the status** of deployed endpoints, checkpoints, and running containers\u002Fdeployments\n- User wants to explore or deploy applications from the Dell app catalog\n- User needs to check system compatibility with Dell AI platforms\n- User asks about Dell AI, Dell Enterprise Hub, or DEH\n- User wants to authenticate with Dell AI using a Hugging Face token\n- User wants to manage local\u002Fglobal **environment variables** for dell-ai\n- User wants to check their system info or check system compatibility against Dell AI validated configs\n\n## Installation\n\n```bash\nuv pip install dell-ai\n```\n\n## Authentication\n\nUses a Hugging Face token. Auto-loads from HF token cache, or accepts one directly.\n\n```python\nfrom dell_ai import DellAIClient\n\nclient = DellAIClient()                        # Uses cached HF token\nclient = DellAIClient(token=\"your_hf_token\")   # Or explicit token\n\nclient.is_authenticated()  # Returns bool\nclient.get_user_info()     # Returns dict with name, email, orgs, etc.\n```\n\n## Models\n\n### Search and list models\n\n`list_models` returns `List[str]` of IDs. `search_models` accepts the same filters but returns full `Model` objects.\n\n```python\n# List model IDs (all filters are optional and combinable)\nmodels = client.list_models(\n    query=\"llama\",                    # Search name\u002Fdescription\n    multimodal=True,                  # True for multimodal, False for text-only\n    min_size=7000,                    # Min params in millions\n    max_size=70000,                   # Max params in millions\n    license_filter=\"apache\",          # License substring match\n    platform_id=\"xe9680-nvidia-h200\", # Only models compatible with this platform\n)\n\n# Full Model objects with same filters\nresults = client.search_models(query=\"llama\", multimodal=True)\n# Model fields: repo_name, description, license, size, is_multimodal,\n#               has_system_prompt, creator_type, status, configs_deploy\n```\n\n### Get model details and compatible platforms\n\n```python\n# Returns Model object\nmodel = client.get_model(\"meta-llama\u002FLlama-4-Maverick-17B-128E-Instruct\")\n\nplatforms = client.get_compatible_platforms(\"google\u002Fgemma-3-27b-it\")\n# Returns List[PlatformCompatibility] with platform_id and configs (List[ModelConfig])\nfor p in platforms:\n    print(p.platform_id, [c.model_dump() for c in p.configs])\n```\n\n### Check model access (gated repos)\n\n```python\n# Returns True, or raises GatedRepoAccessError \u002F AuthenticationError\nclient.check_model_access(\"meta-llama\u002FLlama-4-Maverick-17B-128E-Instruct\")\n```\n\n### Get model Deployment Snippets\n\nAutomatically validates model access and model-platform compatibility.\n\nProvide **either** `num_gpus` (manual sizing) **or** `goodput` (server-optimized) —\nthe two are mutually exclusive, and exactly one is required.\n\n```python\n# Basic deployment snippet with manual GPU sizing\nsnippet = client.get_deployment_snippet(\n    model_id=\"meta-llama\u002FLlama-4-Maverick-17B-128E-Instruct\",\n    platform_id=\"xe9680-nvidia-h200\",\n    engine=\"docker\",       # \"docker\" or \"kubernetes\"\n    num_gpus=8,\n    num_replicas=1,\n)\nprint(snippet)  # Ready-to-use docker command or k8s manifest\n\n# Goodput-optimized: DEH returns the GPU count and optimized params for the\n# scenario. Omit num_gpus when using goodput.\nsnippet = client.get_deployment_snippet(\n    model_id=\"nvidia\u002FMiniMax-M2.7-NVFP4\",\n    platform_id=\"xe9680-nvidia-h200\",\n    engine=\"docker\",\n    goodput=\"balanced\",    # balanced | long-context | high-concurrency | performance\n)\n```\n\nNot every (model, platform, scenario) combination has an optimized config; the\nAPI is the source of truth and raises `ResourceNotFoundError` with a message\nlike `No optimized config for \"balanced\" scenario on this SKU.` when it doesn't.\n\n### Goodput scenarios reference data\n\nGlobal, static reference data: scenario definitions, SLO field docs, and SLO\ntargets per SKU. Cached on disk after the first call.\n\n```python\nref = client.get_goodput_scenarios()\n# GoodputReference fields:\n#   scenarios            -> List[Scenario] (id, label, description)\n#   slo_field_descriptions -> Dict[str, str]\n#   slos_by_sku          -> Dict[SkuId, Dict[scenario, Slo]] (sparse)\n# Slo fields: max_model_context, virtual_users, input_tokens [min,max], output_tokens [min,max]\n\nslo = ref.slos_by_sku.get(\"xe9680-nvidia-h100\", {}).get(\"balanced\")\n```\n\n## Platforms\n\n```python\nplatforms = client.list_platforms()  # Returns List[str] of SKU IDs\n\nplatform = client.get_platform(\"xe9680-nvidia-h200\")\n# Fields: id, name, platform_type, vendor, accelerator_type, accelerator,\n#   gpuram, gpuinterconnect, totalgpucount, product_name\n\nsys_infos = client.get_platform_system_info(\"xe9680-nvidia-h200\")  # List[SystemInfo]\n```\n\n## Applications\n\n### List and inspect apps\n\n```python\napps = client.list_apps()  # Returns List[str] of app names\n\napp = client.get_app(\"openwebui\")\n# App fields: id, name, license, description, features, instructions,\n#   tags, recommendedModels, components\n```\n\n### Explore configurable parameters\n\n```python\nfor component in app.components:\n    for param in component.config:\n        print(f\"  {param.name}: {param.helmPath} ({param.type}), default={param.default}\")\n    for secret in component.secrets:\n        print(f\"  [secret] {secret.name}: {secret.helmPath}\")\n```\n\n### Generate app deployment snippet\n\nEach config item requires: `helmPath`, `type` (`\"string\"`, `\"boolean\"`, `\"number\"`, or `\"json\"`), and `value`.\n\n```python\nconfig = [\n    {\"helmPath\": \"main.config.storageClassName\", \"type\": \"string\", \"value\": \"gp2\"},\n    {\"helmPath\": \"main.config.enableOpenAI\", \"type\": \"boolean\", \"value\": True},\n]\nsnippet = client.get_app_snippet(\"openwebui\", config)\nprint(snippet)  # Helm install command\n```\n\n## Deploying models and applications (local execution)\n\nBeyond generating snippets, dell-ai can **execute** them on the local node using\nthe locally available engine: `docker` (Docker CLI), `kubernetes` (`kubectl\napply`), or Helm (apps). The relevant tool must be installed and configured.\n\nDeployments run **detached (background) by default**. For Docker, dell-ai\nautomatically:\n- converts interactive flags (`-it`) to detached (`-d`) and captures the\n  container ID and inferred endpoint URL,\n- remaps the host port to a free one if the snippet's port is already in use,\n- allocates and pins free GPU indices.\n\nOn success, metadata (endpoint, engine, container\u002Fdeployment ID, assigned GPUs)\nis written to the **deployment registry** (see below).\n\nProvide **either** `num_gpus` **or** `goodput` — mutually exclusive, exactly one\nrequired (same rule as snippet generation).\n\n```python\n# Deploy a model on the local node\nresult = client.deploy_model(\n    model_id=\"meta-llama\u002FLlama-4-Maverick-17B-128E-Instruct\",\n    platform_id=\"xe9680-nvidia-h200\",\n    engine=\"docker\",              # \"docker\" or \"kubernetes\"\n    num_gpus=8,                   # or goodput=\"balanced\" (omit num_gpus)\n    num_replicas=1,\n    detach=True,                  # False = run in foreground\n    # local_dir=\"\u002Fdata\u002Fweights\",           # mount local weights (Docker only)\n    # hf_cache_dir=\"~\u002F.cache\u002Fhuggingface\",  # or reuse an HF cache (mutually exclusive)\n)\n# result dict: success(bool), engine, endpoint, container_id | k8s_deployment, error?\n\n# Deploy an app (Helm)\nresult = client.deploy_app(app_id=\"openwebui\", config=[...], detach=True)\n```\n\n`deploy_model` \u002F `deploy_app` **do not raise on runtime failure** — they return\n`{\"success\": False, \"error\": ...}`. They still raise `ValidationError`,\n`ResourceNotFoundError`, or `GatedRepoAccessError` for bad parameters or access.\n\nCLI equivalents:\n\n```bash\n# Manual GPU sizing \u002F goodput\ndell-ai models deploy -m \u003Cmodel_id> -p \u003Cplatform_id> -e docker --gpus 8\ndell-ai models deploy -m \u003Cmodel_id> -p \u003Cplatform_id> -e docker --goodput balanced\n\n# Kubernetes \u002F foreground\ndell-ai models deploy -m \u003Cmodel_id> -p \u003Cplatform_id> -e kubernetes --gpus 8\ndell-ai models deploy -m \u003Cmodel_id> -p \u003Cplatform_id> -e docker --no-detach\n\n# Serve local weights instead of downloading (Docker only; path must exist; mutually exclusive)\ndell-ai models deploy -m \u003Cmodel_id> -p \u003Cplatform_id> -e docker --gpus 8 --local-dir \u002Fdata\u002Fweights\ndell-ai models deploy -m \u003Cmodel_id> -p \u003Cplatform_id> -e docker --gpus 8 --hf-cache-dir ~\u002F.cache\u002Fhuggingface\n\n# App (Helm)\ndell-ai apps deploy openwebui --config '{\"config\":[{\"helmPath\":\"main.config.storageClassName\",\"type\":\"string\",\"value\":\"gp2\"}]}'\n```\n\n## Deployment registry and teardown\n\nSuccessful deployments are tracked in a registry so they can be listed,\ninspected, and torn down. Two scopes:\n- **Local** — `.dell-ai-deployments.json` (current working directory)\n- **Global** — `~\u002F.config\u002Fdell-ai\u002Fdeployments.json`\n\nEach entry is keyed by a **deployment ID** (defaults to the model ID; additional\ninstances of the same model are suffixed, e.g. `org\u002Fmodel_1`, each with its own\nhost port and GPU indices). Running DEH Docker containers are auto-discovered and\nadded; entries whose containers are gone are pruned automatically.\n\n```python\nfrom dell_ai import deployments\n\ndeployments.list_deployments()             # dict keyed by deployment ID\ndeployments.get_deployment(\"org\u002Fmodel_1\")    # single entry or None\ndeployments.delete_deployment(\"org\u002Fmodel_1\") # remove registry entry (no container stop)\n```\n\nTear down a deployment — stops the Docker container \u002F Kubernetes deployment **and**\nremoves its registry entry:\n\n```bash\ndell-ai models undeploy -d \u003Cdeployment_id>\n```\n\n`undeploy` acts on **one deployment at a time**. Always run\n`dell-ai status` first to see if the target deployment ID exists.\n\n## Environment variables\n\ndell-ai persists configuration as environment variables in two scopes,\nauto-loaded into `os.environ` on CLI startup and `DellAIClient` init. Resolution\nprecedence: **shell env → local → global**.\n- **Local** — `.dell-ai-env.json` (current working directory)\n- **Global** — `~\u002F.config\u002Fdell-ai\u002Fenv.json`\n\n```python\nfrom dell_ai import env\n\nenv.set_env_var(\"DELL_AI_CHECKPOINT\", \"\u002Fdata\u002Fckpt\", is_global=False)\nenv.get_env_var(\"DELL_AI_CHECKPOINT\")\nenv.list_env_vars()   # is_global: None=combined, True=global only, False=local only\nenv.delete_env_var(\"DELL_AI_CHECKPOINT\", is_global=False)\n```\n\n```bash\ndell-ai env set DELL_AI_CHECKPOINT \u002Fdata\u002Fckpt   # add --global for the global scope\ndell-ai env get DELL_AI_CHECKPOINT\ndell-ai env list                                # --local \u002F --global to scope\ndell-ai env delete DELL_AI_CHECKPOINT           # --global to delete from global\n```\n\n## Checking deployment status\n\n```bash\ndell-ai status          # active deployments, checkpoints, running Docker\u002FK8s\ndell-ai status --clean  # also remove exited containers and stopped K8s deployments\n```\n\nReports:\n- **Active deployments** — reads the deployment registry, probes each endpoint\n  (online + response time); auto-discovers running DEH containers.\n- **Checkpoints** — existence, type, and size of paths in `DELL_AI_*_CHECKPOINT`\n  environment variables.\n- **Active Docker\u002FK8s** — running DEH containers and Kubernetes deployments.\n\nDocker\u002FKubernetes scanning is skipped gracefully when `docker` \u002F `kubectl` are\nnot available on the node.\n\n## System Utilities (CLI only, Linux)\n\n```bash\ndell-ai utils describe-system              # Get system info as JSON\ndell-ai utils describe-system -o out.json  # Save to file\ndell-ai utils describe-system | jq         # Pretty-print\ndell-ai utils check-system                 # Check compatibility against Dell AI validated configs\n```\n\nRequires: `lscpu`, `lspci`, `lsblk`, `dmidecode`. For GPUs: `nvidia-smi`, `nvidia-ctk`. For K8s: `kubectl`.\n\n## Exceptions\n\nAll in `dell_ai.exceptions`:\n- `AuthenticationError` — Invalid or missing token\n- `GatedRepoAccessError` — No access to a gated model repository\n- `ResourceNotFoundError` — Model, platform, or app not found\n- `ValidationError` — Invalid parameters (may include `valid_values`)\n- `APIError` — General API errors\n",{"data":36,"body":37},{"name":4,"description":6},{"type":38,"children":39},"root",[40,48,54,61,163,169,209,215,220,290,296,303,338,462,468,530,536,559,565,570,604,754,775,781,786,856,862,923,929,935,981,987,1034,1040,1099,1154,1160,1195,1207,1241,1253,1280,1402,1457,1462,2027,2033,2038,2074,2094,2140,2152,2198,2224,2230,2257,2289,2343,2449,2455,2500,2505,2546,2565,2571,2681,2736,2742,2755,2818],{"type":41,"tag":42,"props":43,"children":44},"element","h1",{"id":4},[45],{"type":46,"value":47},"text","Dell AI",{"type":41,"tag":49,"props":50,"children":51},"p",{},[52],{"type":46,"value":53},"dell-ai interacts with the Dell Enterprise Hub (DEH) — discovering models, exploring hardware platforms, generating deployment snippets, managing app catalog entries, checking system compatibility, etc.",{"type":41,"tag":55,"props":56,"children":58},"h2",{"id":57},"when-to-use",[59],{"type":46,"value":60},"When to use",{"type":41,"tag":62,"props":63,"children":64},"ul",{},[65,71,76,81,86,91,104,115,126,131,136,141,146,158],{"type":41,"tag":66,"props":67,"children":68},"li",{},[69],{"type":46,"value":70},"User wants to find AI models available on Dell Enterprise Hub",{"type":41,"tag":66,"props":72,"children":73},{},[74],{"type":46,"value":75},"User wants to find AI models available on DEH for their hardware platform",{"type":41,"tag":66,"props":77,"children":78},{},[79],{"type":46,"value":80},"User needs to check which Dell hardware platforms support a given model",{"type":41,"tag":66,"props":82,"children":83},{},[84],{"type":46,"value":85},"User wants to generate Docker or Kubernetes deployment commands for a model and then run it.",{"type":41,"tag":66,"props":87,"children":88},{},[89],{"type":46,"value":90},"User wants a goodput-optimized deployment snippet (balanced, long-context, high-concurrency, performance).",{"type":41,"tag":66,"props":92,"children":93},{},[94,96,102],{"type":46,"value":95},"User wants to ",{"type":41,"tag":97,"props":98,"children":99},"strong",{},[100],{"type":46,"value":101},"deploy",{"type":46,"value":103}," a model or app directly onto the local node (not just generate a snippet)",{"type":41,"tag":66,"props":105,"children":106},{},[107,108,113],{"type":46,"value":95},{"type":41,"tag":97,"props":109,"children":110},{},[111],{"type":46,"value":112},"tear down",{"type":46,"value":114}," \u002F undeploy a running deployment",{"type":41,"tag":66,"props":116,"children":117},{},[118,119,124],{"type":46,"value":95},{"type":41,"tag":97,"props":120,"children":121},{},[122],{"type":46,"value":123},"check the status",{"type":46,"value":125}," of deployed endpoints, checkpoints, and running containers\u002Fdeployments",{"type":41,"tag":66,"props":127,"children":128},{},[129],{"type":46,"value":130},"User wants to explore or deploy applications from the Dell app catalog",{"type":41,"tag":66,"props":132,"children":133},{},[134],{"type":46,"value":135},"User needs to check system compatibility with Dell AI platforms",{"type":41,"tag":66,"props":137,"children":138},{},[139],{"type":46,"value":140},"User asks about Dell AI, Dell Enterprise Hub, or DEH",{"type":41,"tag":66,"props":142,"children":143},{},[144],{"type":46,"value":145},"User wants to authenticate with Dell AI using a Hugging Face token",{"type":41,"tag":66,"props":147,"children":148},{},[149,151,156],{"type":46,"value":150},"User wants to manage local\u002Fglobal ",{"type":41,"tag":97,"props":152,"children":153},{},[154],{"type":46,"value":155},"environment variables",{"type":46,"value":157}," for dell-ai",{"type":41,"tag":66,"props":159,"children":160},{},[161],{"type":46,"value":162},"User wants to check their system info or check system compatibility against Dell AI validated configs",{"type":41,"tag":55,"props":164,"children":166},{"id":165},"installation",[167],{"type":46,"value":168},"Installation",{"type":41,"tag":170,"props":171,"children":176},"pre",{"className":172,"code":173,"language":174,"meta":175,"style":175},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","uv pip install dell-ai\n","bash","",[177],{"type":41,"tag":178,"props":179,"children":180},"code",{"__ignoreMap":175},[181],{"type":41,"tag":182,"props":183,"children":186},"span",{"class":184,"line":185},"line",1,[187,193,199,204],{"type":41,"tag":182,"props":188,"children":190},{"style":189},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[191],{"type":46,"value":192},"uv",{"type":41,"tag":182,"props":194,"children":196},{"style":195},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[197],{"type":46,"value":198}," pip",{"type":41,"tag":182,"props":200,"children":201},{"style":195},[202],{"type":46,"value":203}," install",{"type":41,"tag":182,"props":205,"children":206},{"style":195},[207],{"type":46,"value":208}," dell-ai\n",{"type":41,"tag":55,"props":210,"children":212},{"id":211},"authentication",[213],{"type":46,"value":214},"Authentication",{"type":41,"tag":49,"props":216,"children":217},{},[218],{"type":46,"value":219},"Uses a Hugging Face token. Auto-loads from HF token cache, or accepts one directly.",{"type":41,"tag":170,"props":221,"children":225},{"className":222,"code":223,"language":224,"meta":175,"style":175},"language-python shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","from dell_ai import DellAIClient\n\nclient = DellAIClient()                        # Uses cached HF token\nclient = DellAIClient(token=\"your_hf_token\")   # Or explicit token\n\nclient.is_authenticated()  # Returns bool\nclient.get_user_info()     # Returns dict with name, email, orgs, etc.\n","python",[226],{"type":41,"tag":178,"props":227,"children":228},{"__ignoreMap":175},[229,237,247,256,265,273,282],{"type":41,"tag":182,"props":230,"children":231},{"class":184,"line":185},[232],{"type":41,"tag":182,"props":233,"children":234},{},[235],{"type":46,"value":236},"from dell_ai import DellAIClient\n",{"type":41,"tag":182,"props":238,"children":240},{"class":184,"line":239},2,[241],{"type":41,"tag":182,"props":242,"children":244},{"emptyLinePlaceholder":243},true,[245],{"type":46,"value":246},"\n",{"type":41,"tag":182,"props":248,"children":250},{"class":184,"line":249},3,[251],{"type":41,"tag":182,"props":252,"children":253},{},[254],{"type":46,"value":255},"client = DellAIClient()                        # Uses cached HF token\n",{"type":41,"tag":182,"props":257,"children":259},{"class":184,"line":258},4,[260],{"type":41,"tag":182,"props":261,"children":262},{},[263],{"type":46,"value":264},"client = DellAIClient(token=\"your_hf_token\")   # Or explicit token\n",{"type":41,"tag":182,"props":266,"children":268},{"class":184,"line":267},5,[269],{"type":41,"tag":182,"props":270,"children":271},{"emptyLinePlaceholder":243},[272],{"type":46,"value":246},{"type":41,"tag":182,"props":274,"children":276},{"class":184,"line":275},6,[277],{"type":41,"tag":182,"props":278,"children":279},{},[280],{"type":46,"value":281},"client.is_authenticated()  # Returns bool\n",{"type":41,"tag":182,"props":283,"children":284},{"class":184,"line":28},[285],{"type":41,"tag":182,"props":286,"children":287},{},[288],{"type":46,"value":289},"client.get_user_info()     # Returns dict with name, email, orgs, etc.\n",{"type":41,"tag":55,"props":291,"children":293},{"id":292},"models",[294],{"type":46,"value":295},"Models",{"type":41,"tag":297,"props":298,"children":300},"h3",{"id":299},"search-and-list-models",[301],{"type":46,"value":302},"Search and list models",{"type":41,"tag":49,"props":304,"children":305},{},[306,312,314,320,322,328,330,336],{"type":41,"tag":178,"props":307,"children":309},{"className":308},[],[310],{"type":46,"value":311},"list_models",{"type":46,"value":313}," returns ",{"type":41,"tag":178,"props":315,"children":317},{"className":316},[],[318],{"type":46,"value":319},"List[str]",{"type":46,"value":321}," of IDs. ",{"type":41,"tag":178,"props":323,"children":325},{"className":324},[],[326],{"type":46,"value":327},"search_models",{"type":46,"value":329}," accepts the same filters but returns full ",{"type":41,"tag":178,"props":331,"children":333},{"className":332},[],[334],{"type":46,"value":335},"Model",{"type":46,"value":337}," objects.",{"type":41,"tag":170,"props":339,"children":341},{"className":222,"code":340,"language":224,"meta":175,"style":175},"# List model IDs (all filters are optional and combinable)\nmodels = client.list_models(\n    query=\"llama\",                    # Search name\u002Fdescription\n    multimodal=True,                  # True for multimodal, False for text-only\n    min_size=7000,                    # Min params in millions\n    max_size=70000,                   # Max params in millions\n    license_filter=\"apache\",          # License substring match\n    platform_id=\"xe9680-nvidia-h200\", # Only models compatible with this platform\n)\n\n# Full Model objects with same filters\nresults = client.search_models(query=\"llama\", multimodal=True)\n# Model fields: repo_name, description, license, size, is_multimodal,\n#               has_system_prompt, creator_type, status, configs_deploy\n",[342],{"type":41,"tag":178,"props":343,"children":344},{"__ignoreMap":175},[345,353,361,369,377,385,393,401,410,419,426,435,444,453],{"type":41,"tag":182,"props":346,"children":347},{"class":184,"line":185},[348],{"type":41,"tag":182,"props":349,"children":350},{},[351],{"type":46,"value":352},"# List model IDs (all filters are optional and combinable)\n",{"type":41,"tag":182,"props":354,"children":355},{"class":184,"line":239},[356],{"type":41,"tag":182,"props":357,"children":358},{},[359],{"type":46,"value":360},"models = client.list_models(\n",{"type":41,"tag":182,"props":362,"children":363},{"class":184,"line":249},[364],{"type":41,"tag":182,"props":365,"children":366},{},[367],{"type":46,"value":368},"    query=\"llama\",                    # Search name\u002Fdescription\n",{"type":41,"tag":182,"props":370,"children":371},{"class":184,"line":258},[372],{"type":41,"tag":182,"props":373,"children":374},{},[375],{"type":46,"value":376},"    multimodal=True,                  # True for multimodal, False for text-only\n",{"type":41,"tag":182,"props":378,"children":379},{"class":184,"line":267},[380],{"type":41,"tag":182,"props":381,"children":382},{},[383],{"type":46,"value":384},"    min_size=7000,                    # Min params in millions\n",{"type":41,"tag":182,"props":386,"children":387},{"class":184,"line":275},[388],{"type":41,"tag":182,"props":389,"children":390},{},[391],{"type":46,"value":392},"    max_size=70000,                   # Max params in millions\n",{"type":41,"tag":182,"props":394,"children":395},{"class":184,"line":28},[396],{"type":41,"tag":182,"props":397,"children":398},{},[399],{"type":46,"value":400},"    license_filter=\"apache\",          # License substring match\n",{"type":41,"tag":182,"props":402,"children":404},{"class":184,"line":403},8,[405],{"type":41,"tag":182,"props":406,"children":407},{},[408],{"type":46,"value":409},"    platform_id=\"xe9680-nvidia-h200\", # Only models compatible with this platform\n",{"type":41,"tag":182,"props":411,"children":413},{"class":184,"line":412},9,[414],{"type":41,"tag":182,"props":415,"children":416},{},[417],{"type":46,"value":418},")\n",{"type":41,"tag":182,"props":420,"children":421},{"class":184,"line":24},[422],{"type":41,"tag":182,"props":423,"children":424},{"emptyLinePlaceholder":243},[425],{"type":46,"value":246},{"type":41,"tag":182,"props":427,"children":429},{"class":184,"line":428},11,[430],{"type":41,"tag":182,"props":431,"children":432},{},[433],{"type":46,"value":434},"# Full Model objects with same filters\n",{"type":41,"tag":182,"props":436,"children":438},{"class":184,"line":437},12,[439],{"type":41,"tag":182,"props":440,"children":441},{},[442],{"type":46,"value":443},"results = client.search_models(query=\"llama\", multimodal=True)\n",{"type":41,"tag":182,"props":445,"children":447},{"class":184,"line":446},13,[448],{"type":41,"tag":182,"props":449,"children":450},{},[451],{"type":46,"value":452},"# Model fields: repo_name, description, license, size, is_multimodal,\n",{"type":41,"tag":182,"props":454,"children":456},{"class":184,"line":455},14,[457],{"type":41,"tag":182,"props":458,"children":459},{},[460],{"type":46,"value":461},"#               has_system_prompt, creator_type, status, configs_deploy\n",{"type":41,"tag":297,"props":463,"children":465},{"id":464},"get-model-details-and-compatible-platforms",[466],{"type":46,"value":467},"Get model details and compatible platforms",{"type":41,"tag":170,"props":469,"children":471},{"className":222,"code":470,"language":224,"meta":175,"style":175},"# Returns Model object\nmodel = client.get_model(\"meta-llama\u002FLlama-4-Maverick-17B-128E-Instruct\")\n\nplatforms = client.get_compatible_platforms(\"google\u002Fgemma-3-27b-it\")\n# Returns List[PlatformCompatibility] with platform_id and configs (List[ModelConfig])\nfor p in platforms:\n    print(p.platform_id, [c.model_dump() for c in p.configs])\n",[472],{"type":41,"tag":178,"props":473,"children":474},{"__ignoreMap":175},[475,483,491,498,506,514,522],{"type":41,"tag":182,"props":476,"children":477},{"class":184,"line":185},[478],{"type":41,"tag":182,"props":479,"children":480},{},[481],{"type":46,"value":482},"# Returns Model object\n",{"type":41,"tag":182,"props":484,"children":485},{"class":184,"line":239},[486],{"type":41,"tag":182,"props":487,"children":488},{},[489],{"type":46,"value":490},"model = client.get_model(\"meta-llama\u002FLlama-4-Maverick-17B-128E-Instruct\")\n",{"type":41,"tag":182,"props":492,"children":493},{"class":184,"line":249},[494],{"type":41,"tag":182,"props":495,"children":496},{"emptyLinePlaceholder":243},[497],{"type":46,"value":246},{"type":41,"tag":182,"props":499,"children":500},{"class":184,"line":258},[501],{"type":41,"tag":182,"props":502,"children":503},{},[504],{"type":46,"value":505},"platforms = client.get_compatible_platforms(\"google\u002Fgemma-3-27b-it\")\n",{"type":41,"tag":182,"props":507,"children":508},{"class":184,"line":267},[509],{"type":41,"tag":182,"props":510,"children":511},{},[512],{"type":46,"value":513},"# Returns List[PlatformCompatibility] with platform_id and configs (List[ModelConfig])\n",{"type":41,"tag":182,"props":515,"children":516},{"class":184,"line":275},[517],{"type":41,"tag":182,"props":518,"children":519},{},[520],{"type":46,"value":521},"for p in platforms:\n",{"type":41,"tag":182,"props":523,"children":524},{"class":184,"line":28},[525],{"type":41,"tag":182,"props":526,"children":527},{},[528],{"type":46,"value":529},"    print(p.platform_id, [c.model_dump() for c in p.configs])\n",{"type":41,"tag":297,"props":531,"children":533},{"id":532},"check-model-access-gated-repos",[534],{"type":46,"value":535},"Check model access (gated repos)",{"type":41,"tag":170,"props":537,"children":539},{"className":222,"code":538,"language":224,"meta":175,"style":175},"# Returns True, or raises GatedRepoAccessError \u002F AuthenticationError\nclient.check_model_access(\"meta-llama\u002FLlama-4-Maverick-17B-128E-Instruct\")\n",[540],{"type":41,"tag":178,"props":541,"children":542},{"__ignoreMap":175},[543,551],{"type":41,"tag":182,"props":544,"children":545},{"class":184,"line":185},[546],{"type":41,"tag":182,"props":547,"children":548},{},[549],{"type":46,"value":550},"# Returns True, or raises GatedRepoAccessError \u002F AuthenticationError\n",{"type":41,"tag":182,"props":552,"children":553},{"class":184,"line":239},[554],{"type":41,"tag":182,"props":555,"children":556},{},[557],{"type":46,"value":558},"client.check_model_access(\"meta-llama\u002FLlama-4-Maverick-17B-128E-Instruct\")\n",{"type":41,"tag":297,"props":560,"children":562},{"id":561},"get-model-deployment-snippets",[563],{"type":46,"value":564},"Get model Deployment Snippets",{"type":41,"tag":49,"props":566,"children":567},{},[568],{"type":46,"value":569},"Automatically validates model access and model-platform compatibility.",{"type":41,"tag":49,"props":571,"children":572},{},[573,575,580,582,588,590,595,596,602],{"type":46,"value":574},"Provide ",{"type":41,"tag":97,"props":576,"children":577},{},[578],{"type":46,"value":579},"either",{"type":46,"value":581}," ",{"type":41,"tag":178,"props":583,"children":585},{"className":584},[],[586],{"type":46,"value":587},"num_gpus",{"type":46,"value":589}," (manual sizing) ",{"type":41,"tag":97,"props":591,"children":592},{},[593],{"type":46,"value":594},"or",{"type":46,"value":581},{"type":41,"tag":178,"props":597,"children":599},{"className":598},[],[600],{"type":46,"value":601},"goodput",{"type":46,"value":603}," (server-optimized) —\nthe two are mutually exclusive, and exactly one is required.",{"type":41,"tag":170,"props":605,"children":607},{"className":222,"code":606,"language":224,"meta":175,"style":175},"# Basic deployment snippet with manual GPU sizing\nsnippet = client.get_deployment_snippet(\n    model_id=\"meta-llama\u002FLlama-4-Maverick-17B-128E-Instruct\",\n    platform_id=\"xe9680-nvidia-h200\",\n    engine=\"docker\",       # \"docker\" or \"kubernetes\"\n    num_gpus=8,\n    num_replicas=1,\n)\nprint(snippet)  # Ready-to-use docker command or k8s manifest\n\n# Goodput-optimized: DEH returns the GPU count and optimized params for the\n# scenario. Omit num_gpus when using goodput.\nsnippet = client.get_deployment_snippet(\n    model_id=\"nvidia\u002FMiniMax-M2.7-NVFP4\",\n    platform_id=\"xe9680-nvidia-h200\",\n    engine=\"docker\",\n    goodput=\"balanced\",    # balanced | long-context | high-concurrency | performance\n)\n",[608],{"type":41,"tag":178,"props":609,"children":610},{"__ignoreMap":175},[611,619,627,635,643,651,659,667,674,682,689,697,705,712,720,728,737,746],{"type":41,"tag":182,"props":612,"children":613},{"class":184,"line":185},[614],{"type":41,"tag":182,"props":615,"children":616},{},[617],{"type":46,"value":618},"# Basic deployment snippet with manual GPU sizing\n",{"type":41,"tag":182,"props":620,"children":621},{"class":184,"line":239},[622],{"type":41,"tag":182,"props":623,"children":624},{},[625],{"type":46,"value":626},"snippet = client.get_deployment_snippet(\n",{"type":41,"tag":182,"props":628,"children":629},{"class":184,"line":249},[630],{"type":41,"tag":182,"props":631,"children":632},{},[633],{"type":46,"value":634},"    model_id=\"meta-llama\u002FLlama-4-Maverick-17B-128E-Instruct\",\n",{"type":41,"tag":182,"props":636,"children":637},{"class":184,"line":258},[638],{"type":41,"tag":182,"props":639,"children":640},{},[641],{"type":46,"value":642},"    platform_id=\"xe9680-nvidia-h200\",\n",{"type":41,"tag":182,"props":644,"children":645},{"class":184,"line":267},[646],{"type":41,"tag":182,"props":647,"children":648},{},[649],{"type":46,"value":650},"    engine=\"docker\",       # \"docker\" or \"kubernetes\"\n",{"type":41,"tag":182,"props":652,"children":653},{"class":184,"line":275},[654],{"type":41,"tag":182,"props":655,"children":656},{},[657],{"type":46,"value":658},"    num_gpus=8,\n",{"type":41,"tag":182,"props":660,"children":661},{"class":184,"line":28},[662],{"type":41,"tag":182,"props":663,"children":664},{},[665],{"type":46,"value":666},"    num_replicas=1,\n",{"type":41,"tag":182,"props":668,"children":669},{"class":184,"line":403},[670],{"type":41,"tag":182,"props":671,"children":672},{},[673],{"type":46,"value":418},{"type":41,"tag":182,"props":675,"children":676},{"class":184,"line":412},[677],{"type":41,"tag":182,"props":678,"children":679},{},[680],{"type":46,"value":681},"print(snippet)  # Ready-to-use docker command or k8s manifest\n",{"type":41,"tag":182,"props":683,"children":684},{"class":184,"line":24},[685],{"type":41,"tag":182,"props":686,"children":687},{"emptyLinePlaceholder":243},[688],{"type":46,"value":246},{"type":41,"tag":182,"props":690,"children":691},{"class":184,"line":428},[692],{"type":41,"tag":182,"props":693,"children":694},{},[695],{"type":46,"value":696},"# Goodput-optimized: DEH returns the GPU count and optimized params for the\n",{"type":41,"tag":182,"props":698,"children":699},{"class":184,"line":437},[700],{"type":41,"tag":182,"props":701,"children":702},{},[703],{"type":46,"value":704},"# scenario. Omit num_gpus when using goodput.\n",{"type":41,"tag":182,"props":706,"children":707},{"class":184,"line":446},[708],{"type":41,"tag":182,"props":709,"children":710},{},[711],{"type":46,"value":626},{"type":41,"tag":182,"props":713,"children":714},{"class":184,"line":455},[715],{"type":41,"tag":182,"props":716,"children":717},{},[718],{"type":46,"value":719},"    model_id=\"nvidia\u002FMiniMax-M2.7-NVFP4\",\n",{"type":41,"tag":182,"props":721,"children":723},{"class":184,"line":722},15,[724],{"type":41,"tag":182,"props":725,"children":726},{},[727],{"type":46,"value":642},{"type":41,"tag":182,"props":729,"children":731},{"class":184,"line":730},16,[732],{"type":41,"tag":182,"props":733,"children":734},{},[735],{"type":46,"value":736},"    engine=\"docker\",\n",{"type":41,"tag":182,"props":738,"children":740},{"class":184,"line":739},17,[741],{"type":41,"tag":182,"props":742,"children":743},{},[744],{"type":46,"value":745},"    goodput=\"balanced\",    # balanced | long-context | high-concurrency | performance\n",{"type":41,"tag":182,"props":747,"children":749},{"class":184,"line":748},18,[750],{"type":41,"tag":182,"props":751,"children":752},{},[753],{"type":46,"value":418},{"type":41,"tag":49,"props":755,"children":756},{},[757,759,765,767,773],{"type":46,"value":758},"Not every (model, platform, scenario) combination has an optimized config; the\nAPI is the source of truth and raises ",{"type":41,"tag":178,"props":760,"children":762},{"className":761},[],[763],{"type":46,"value":764},"ResourceNotFoundError",{"type":46,"value":766}," with a message\nlike ",{"type":41,"tag":178,"props":768,"children":770},{"className":769},[],[771],{"type":46,"value":772},"No optimized config for \"balanced\" scenario on this SKU.",{"type":46,"value":774}," when it doesn't.",{"type":41,"tag":297,"props":776,"children":778},{"id":777},"goodput-scenarios-reference-data",[779],{"type":46,"value":780},"Goodput scenarios reference data",{"type":41,"tag":49,"props":782,"children":783},{},[784],{"type":46,"value":785},"Global, static reference data: scenario definitions, SLO field docs, and SLO\ntargets per SKU. Cached on disk after the first call.",{"type":41,"tag":170,"props":787,"children":789},{"className":222,"code":788,"language":224,"meta":175,"style":175},"ref = client.get_goodput_scenarios()\n# GoodputReference fields:\n#   scenarios            -> List[Scenario] (id, label, description)\n#   slo_field_descriptions -> Dict[str, str]\n#   slos_by_sku          -> Dict[SkuId, Dict[scenario, Slo]] (sparse)\n# Slo fields: max_model_context, virtual_users, input_tokens [min,max], output_tokens [min,max]\n\nslo = ref.slos_by_sku.get(\"xe9680-nvidia-h100\", {}).get(\"balanced\")\n",[790],{"type":41,"tag":178,"props":791,"children":792},{"__ignoreMap":175},[793,801,809,817,825,833,841,848],{"type":41,"tag":182,"props":794,"children":795},{"class":184,"line":185},[796],{"type":41,"tag":182,"props":797,"children":798},{},[799],{"type":46,"value":800},"ref = client.get_goodput_scenarios()\n",{"type":41,"tag":182,"props":802,"children":803},{"class":184,"line":239},[804],{"type":41,"tag":182,"props":805,"children":806},{},[807],{"type":46,"value":808},"# GoodputReference fields:\n",{"type":41,"tag":182,"props":810,"children":811},{"class":184,"line":249},[812],{"type":41,"tag":182,"props":813,"children":814},{},[815],{"type":46,"value":816},"#   scenarios            -> List[Scenario] (id, label, description)\n",{"type":41,"tag":182,"props":818,"children":819},{"class":184,"line":258},[820],{"type":41,"tag":182,"props":821,"children":822},{},[823],{"type":46,"value":824},"#   slo_field_descriptions -> Dict[str, str]\n",{"type":41,"tag":182,"props":826,"children":827},{"class":184,"line":267},[828],{"type":41,"tag":182,"props":829,"children":830},{},[831],{"type":46,"value":832},"#   slos_by_sku          -> Dict[SkuId, Dict[scenario, Slo]] (sparse)\n",{"type":41,"tag":182,"props":834,"children":835},{"class":184,"line":275},[836],{"type":41,"tag":182,"props":837,"children":838},{},[839],{"type":46,"value":840},"# Slo fields: max_model_context, virtual_users, input_tokens [min,max], output_tokens [min,max]\n",{"type":41,"tag":182,"props":842,"children":843},{"class":184,"line":28},[844],{"type":41,"tag":182,"props":845,"children":846},{"emptyLinePlaceholder":243},[847],{"type":46,"value":246},{"type":41,"tag":182,"props":849,"children":850},{"class":184,"line":403},[851],{"type":41,"tag":182,"props":852,"children":853},{},[854],{"type":46,"value":855},"slo = ref.slos_by_sku.get(\"xe9680-nvidia-h100\", {}).get(\"balanced\")\n",{"type":41,"tag":55,"props":857,"children":859},{"id":858},"platforms",[860],{"type":46,"value":861},"Platforms",{"type":41,"tag":170,"props":863,"children":865},{"className":222,"code":864,"language":224,"meta":175,"style":175},"platforms = client.list_platforms()  # Returns List[str] of SKU IDs\n\nplatform = client.get_platform(\"xe9680-nvidia-h200\")\n# Fields: id, name, platform_type, vendor, accelerator_type, accelerator,\n#   gpuram, gpuinterconnect, totalgpucount, product_name\n\nsys_infos = client.get_platform_system_info(\"xe9680-nvidia-h200\")  # List[SystemInfo]\n",[866],{"type":41,"tag":178,"props":867,"children":868},{"__ignoreMap":175},[869,877,884,892,900,908,915],{"type":41,"tag":182,"props":870,"children":871},{"class":184,"line":185},[872],{"type":41,"tag":182,"props":873,"children":874},{},[875],{"type":46,"value":876},"platforms = client.list_platforms()  # Returns List[str] of SKU IDs\n",{"type":41,"tag":182,"props":878,"children":879},{"class":184,"line":239},[880],{"type":41,"tag":182,"props":881,"children":882},{"emptyLinePlaceholder":243},[883],{"type":46,"value":246},{"type":41,"tag":182,"props":885,"children":886},{"class":184,"line":249},[887],{"type":41,"tag":182,"props":888,"children":889},{},[890],{"type":46,"value":891},"platform = client.get_platform(\"xe9680-nvidia-h200\")\n",{"type":41,"tag":182,"props":893,"children":894},{"class":184,"line":258},[895],{"type":41,"tag":182,"props":896,"children":897},{},[898],{"type":46,"value":899},"# Fields: id, name, platform_type, vendor, accelerator_type, accelerator,\n",{"type":41,"tag":182,"props":901,"children":902},{"class":184,"line":267},[903],{"type":41,"tag":182,"props":904,"children":905},{},[906],{"type":46,"value":907},"#   gpuram, gpuinterconnect, totalgpucount, product_name\n",{"type":41,"tag":182,"props":909,"children":910},{"class":184,"line":275},[911],{"type":41,"tag":182,"props":912,"children":913},{"emptyLinePlaceholder":243},[914],{"type":46,"value":246},{"type":41,"tag":182,"props":916,"children":917},{"class":184,"line":28},[918],{"type":41,"tag":182,"props":919,"children":920},{},[921],{"type":46,"value":922},"sys_infos = client.get_platform_system_info(\"xe9680-nvidia-h200\")  # List[SystemInfo]\n",{"type":41,"tag":55,"props":924,"children":926},{"id":925},"applications",[927],{"type":46,"value":928},"Applications",{"type":41,"tag":297,"props":930,"children":932},{"id":931},"list-and-inspect-apps",[933],{"type":46,"value":934},"List and inspect apps",{"type":41,"tag":170,"props":936,"children":938},{"className":222,"code":937,"language":224,"meta":175,"style":175},"apps = client.list_apps()  # Returns List[str] of app names\n\napp = client.get_app(\"openwebui\")\n# App fields: id, name, license, description, features, instructions,\n#   tags, recommendedModels, components\n",[939],{"type":41,"tag":178,"props":940,"children":941},{"__ignoreMap":175},[942,950,957,965,973],{"type":41,"tag":182,"props":943,"children":944},{"class":184,"line":185},[945],{"type":41,"tag":182,"props":946,"children":947},{},[948],{"type":46,"value":949},"apps = client.list_apps()  # Returns List[str] of app names\n",{"type":41,"tag":182,"props":951,"children":952},{"class":184,"line":239},[953],{"type":41,"tag":182,"props":954,"children":955},{"emptyLinePlaceholder":243},[956],{"type":46,"value":246},{"type":41,"tag":182,"props":958,"children":959},{"class":184,"line":249},[960],{"type":41,"tag":182,"props":961,"children":962},{},[963],{"type":46,"value":964},"app = client.get_app(\"openwebui\")\n",{"type":41,"tag":182,"props":966,"children":967},{"class":184,"line":258},[968],{"type":41,"tag":182,"props":969,"children":970},{},[971],{"type":46,"value":972},"# App fields: id, name, license, description, features, instructions,\n",{"type":41,"tag":182,"props":974,"children":975},{"class":184,"line":267},[976],{"type":41,"tag":182,"props":977,"children":978},{},[979],{"type":46,"value":980},"#   tags, recommendedModels, components\n",{"type":41,"tag":297,"props":982,"children":984},{"id":983},"explore-configurable-parameters",[985],{"type":46,"value":986},"Explore configurable parameters",{"type":41,"tag":170,"props":988,"children":990},{"className":222,"code":989,"language":224,"meta":175,"style":175},"for component in app.components:\n    for param in component.config:\n        print(f\"  {param.name}: {param.helmPath} ({param.type}), default={param.default}\")\n    for secret in component.secrets:\n        print(f\"  [secret] {secret.name}: {secret.helmPath}\")\n",[991],{"type":41,"tag":178,"props":992,"children":993},{"__ignoreMap":175},[994,1002,1010,1018,1026],{"type":41,"tag":182,"props":995,"children":996},{"class":184,"line":185},[997],{"type":41,"tag":182,"props":998,"children":999},{},[1000],{"type":46,"value":1001},"for component in app.components:\n",{"type":41,"tag":182,"props":1003,"children":1004},{"class":184,"line":239},[1005],{"type":41,"tag":182,"props":1006,"children":1007},{},[1008],{"type":46,"value":1009},"    for param in component.config:\n",{"type":41,"tag":182,"props":1011,"children":1012},{"class":184,"line":249},[1013],{"type":41,"tag":182,"props":1014,"children":1015},{},[1016],{"type":46,"value":1017},"        print(f\"  {param.name}: {param.helmPath} ({param.type}), default={param.default}\")\n",{"type":41,"tag":182,"props":1019,"children":1020},{"class":184,"line":258},[1021],{"type":41,"tag":182,"props":1022,"children":1023},{},[1024],{"type":46,"value":1025},"    for secret in component.secrets:\n",{"type":41,"tag":182,"props":1027,"children":1028},{"class":184,"line":267},[1029],{"type":41,"tag":182,"props":1030,"children":1031},{},[1032],{"type":46,"value":1033},"        print(f\"  [secret] {secret.name}: {secret.helmPath}\")\n",{"type":41,"tag":297,"props":1035,"children":1037},{"id":1036},"generate-app-deployment-snippet",[1038],{"type":46,"value":1039},"Generate app deployment snippet",{"type":41,"tag":49,"props":1041,"children":1042},{},[1043,1045,1051,1053,1059,1061,1067,1068,1074,1075,1081,1083,1089,1091,1097],{"type":46,"value":1044},"Each config item requires: ",{"type":41,"tag":178,"props":1046,"children":1048},{"className":1047},[],[1049],{"type":46,"value":1050},"helmPath",{"type":46,"value":1052},", ",{"type":41,"tag":178,"props":1054,"children":1056},{"className":1055},[],[1057],{"type":46,"value":1058},"type",{"type":46,"value":1060}," (",{"type":41,"tag":178,"props":1062,"children":1064},{"className":1063},[],[1065],{"type":46,"value":1066},"\"string\"",{"type":46,"value":1052},{"type":41,"tag":178,"props":1069,"children":1071},{"className":1070},[],[1072],{"type":46,"value":1073},"\"boolean\"",{"type":46,"value":1052},{"type":41,"tag":178,"props":1076,"children":1078},{"className":1077},[],[1079],{"type":46,"value":1080},"\"number\"",{"type":46,"value":1082},", or ",{"type":41,"tag":178,"props":1084,"children":1086},{"className":1085},[],[1087],{"type":46,"value":1088},"\"json\"",{"type":46,"value":1090},"), and ",{"type":41,"tag":178,"props":1092,"children":1094},{"className":1093},[],[1095],{"type":46,"value":1096},"value",{"type":46,"value":1098},".",{"type":41,"tag":170,"props":1100,"children":1102},{"className":222,"code":1101,"language":224,"meta":175,"style":175},"config = [\n    {\"helmPath\": \"main.config.storageClassName\", \"type\": \"string\", \"value\": \"gp2\"},\n    {\"helmPath\": \"main.config.enableOpenAI\", \"type\": \"boolean\", \"value\": True},\n]\nsnippet = client.get_app_snippet(\"openwebui\", config)\nprint(snippet)  # Helm install command\n",[1103],{"type":41,"tag":178,"props":1104,"children":1105},{"__ignoreMap":175},[1106,1114,1122,1130,1138,1146],{"type":41,"tag":182,"props":1107,"children":1108},{"class":184,"line":185},[1109],{"type":41,"tag":182,"props":1110,"children":1111},{},[1112],{"type":46,"value":1113},"config = [\n",{"type":41,"tag":182,"props":1115,"children":1116},{"class":184,"line":239},[1117],{"type":41,"tag":182,"props":1118,"children":1119},{},[1120],{"type":46,"value":1121},"    {\"helmPath\": \"main.config.storageClassName\", \"type\": \"string\", \"value\": \"gp2\"},\n",{"type":41,"tag":182,"props":1123,"children":1124},{"class":184,"line":249},[1125],{"type":41,"tag":182,"props":1126,"children":1127},{},[1128],{"type":46,"value":1129},"    {\"helmPath\": \"main.config.enableOpenAI\", \"type\": \"boolean\", \"value\": True},\n",{"type":41,"tag":182,"props":1131,"children":1132},{"class":184,"line":258},[1133],{"type":41,"tag":182,"props":1134,"children":1135},{},[1136],{"type":46,"value":1137},"]\n",{"type":41,"tag":182,"props":1139,"children":1140},{"class":184,"line":267},[1141],{"type":41,"tag":182,"props":1142,"children":1143},{},[1144],{"type":46,"value":1145},"snippet = client.get_app_snippet(\"openwebui\", config)\n",{"type":41,"tag":182,"props":1147,"children":1148},{"class":184,"line":275},[1149],{"type":41,"tag":182,"props":1150,"children":1151},{},[1152],{"type":46,"value":1153},"print(snippet)  # Helm install command\n",{"type":41,"tag":55,"props":1155,"children":1157},{"id":1156},"deploying-models-and-applications-local-execution",[1158],{"type":46,"value":1159},"Deploying models and applications (local execution)",{"type":41,"tag":49,"props":1161,"children":1162},{},[1163,1165,1170,1172,1178,1180,1186,1187,1193],{"type":46,"value":1164},"Beyond generating snippets, dell-ai can ",{"type":41,"tag":97,"props":1166,"children":1167},{},[1168],{"type":46,"value":1169},"execute",{"type":46,"value":1171}," them on the local node using\nthe locally available engine: ",{"type":41,"tag":178,"props":1173,"children":1175},{"className":1174},[],[1176],{"type":46,"value":1177},"docker",{"type":46,"value":1179}," (Docker CLI), ",{"type":41,"tag":178,"props":1181,"children":1183},{"className":1182},[],[1184],{"type":46,"value":1185},"kubernetes",{"type":46,"value":1060},{"type":41,"tag":178,"props":1188,"children":1190},{"className":1189},[],[1191],{"type":46,"value":1192},"kubectl apply",{"type":46,"value":1194},"), or Helm (apps). The relevant tool must be installed and configured.",{"type":41,"tag":49,"props":1196,"children":1197},{},[1198,1200,1205],{"type":46,"value":1199},"Deployments run ",{"type":41,"tag":97,"props":1201,"children":1202},{},[1203],{"type":46,"value":1204},"detached (background) by default",{"type":46,"value":1206},". For Docker, dell-ai\nautomatically:",{"type":41,"tag":62,"props":1208,"children":1209},{},[1210,1231,1236],{"type":41,"tag":66,"props":1211,"children":1212},{},[1213,1215,1221,1223,1229],{"type":46,"value":1214},"converts interactive flags (",{"type":41,"tag":178,"props":1216,"children":1218},{"className":1217},[],[1219],{"type":46,"value":1220},"-it",{"type":46,"value":1222},") to detached (",{"type":41,"tag":178,"props":1224,"children":1226},{"className":1225},[],[1227],{"type":46,"value":1228},"-d",{"type":46,"value":1230},") and captures the\ncontainer ID and inferred endpoint URL,",{"type":41,"tag":66,"props":1232,"children":1233},{},[1234],{"type":46,"value":1235},"remaps the host port to a free one if the snippet's port is already in use,",{"type":41,"tag":66,"props":1237,"children":1238},{},[1239],{"type":46,"value":1240},"allocates and pins free GPU indices.",{"type":41,"tag":49,"props":1242,"children":1243},{},[1244,1246,1251],{"type":46,"value":1245},"On success, metadata (endpoint, engine, container\u002Fdeployment ID, assigned GPUs)\nis written to the ",{"type":41,"tag":97,"props":1247,"children":1248},{},[1249],{"type":46,"value":1250},"deployment registry",{"type":46,"value":1252}," (see below).",{"type":41,"tag":49,"props":1254,"children":1255},{},[1256,1257,1261,1262,1267,1268,1272,1273,1278],{"type":46,"value":574},{"type":41,"tag":97,"props":1258,"children":1259},{},[1260],{"type":46,"value":579},{"type":46,"value":581},{"type":41,"tag":178,"props":1263,"children":1265},{"className":1264},[],[1266],{"type":46,"value":587},{"type":46,"value":581},{"type":41,"tag":97,"props":1269,"children":1270},{},[1271],{"type":46,"value":594},{"type":46,"value":581},{"type":41,"tag":178,"props":1274,"children":1276},{"className":1275},[],[1277],{"type":46,"value":601},{"type":46,"value":1279}," — mutually exclusive, exactly one\nrequired (same rule as snippet generation).",{"type":41,"tag":170,"props":1281,"children":1283},{"className":222,"code":1282,"language":224,"meta":175,"style":175},"# Deploy a model on the local node\nresult = client.deploy_model(\n    model_id=\"meta-llama\u002FLlama-4-Maverick-17B-128E-Instruct\",\n    platform_id=\"xe9680-nvidia-h200\",\n    engine=\"docker\",              # \"docker\" or \"kubernetes\"\n    num_gpus=8,                   # or goodput=\"balanced\" (omit num_gpus)\n    num_replicas=1,\n    detach=True,                  # False = run in foreground\n    # local_dir=\"\u002Fdata\u002Fweights\",           # mount local weights (Docker only)\n    # hf_cache_dir=\"~\u002F.cache\u002Fhuggingface\",  # or reuse an HF cache (mutually exclusive)\n)\n# result dict: success(bool), engine, endpoint, container_id | k8s_deployment, error?\n\n# Deploy an app (Helm)\nresult = client.deploy_app(app_id=\"openwebui\", config=[...], detach=True)\n",[1284],{"type":41,"tag":178,"props":1285,"children":1286},{"__ignoreMap":175},[1287,1295,1303,1310,1317,1325,1333,1340,1348,1356,1364,1371,1379,1386,1394],{"type":41,"tag":182,"props":1288,"children":1289},{"class":184,"line":185},[1290],{"type":41,"tag":182,"props":1291,"children":1292},{},[1293],{"type":46,"value":1294},"# Deploy a model on the local node\n",{"type":41,"tag":182,"props":1296,"children":1297},{"class":184,"line":239},[1298],{"type":41,"tag":182,"props":1299,"children":1300},{},[1301],{"type":46,"value":1302},"result = client.deploy_model(\n",{"type":41,"tag":182,"props":1304,"children":1305},{"class":184,"line":249},[1306],{"type":41,"tag":182,"props":1307,"children":1308},{},[1309],{"type":46,"value":634},{"type":41,"tag":182,"props":1311,"children":1312},{"class":184,"line":258},[1313],{"type":41,"tag":182,"props":1314,"children":1315},{},[1316],{"type":46,"value":642},{"type":41,"tag":182,"props":1318,"children":1319},{"class":184,"line":267},[1320],{"type":41,"tag":182,"props":1321,"children":1322},{},[1323],{"type":46,"value":1324},"    engine=\"docker\",              # \"docker\" or \"kubernetes\"\n",{"type":41,"tag":182,"props":1326,"children":1327},{"class":184,"line":275},[1328],{"type":41,"tag":182,"props":1329,"children":1330},{},[1331],{"type":46,"value":1332},"    num_gpus=8,                   # or goodput=\"balanced\" (omit num_gpus)\n",{"type":41,"tag":182,"props":1334,"children":1335},{"class":184,"line":28},[1336],{"type":41,"tag":182,"props":1337,"children":1338},{},[1339],{"type":46,"value":666},{"type":41,"tag":182,"props":1341,"children":1342},{"class":184,"line":403},[1343],{"type":41,"tag":182,"props":1344,"children":1345},{},[1346],{"type":46,"value":1347},"    detach=True,                  # False = run in foreground\n",{"type":41,"tag":182,"props":1349,"children":1350},{"class":184,"line":412},[1351],{"type":41,"tag":182,"props":1352,"children":1353},{},[1354],{"type":46,"value":1355},"    # local_dir=\"\u002Fdata\u002Fweights\",           # mount local weights (Docker only)\n",{"type":41,"tag":182,"props":1357,"children":1358},{"class":184,"line":24},[1359],{"type":41,"tag":182,"props":1360,"children":1361},{},[1362],{"type":46,"value":1363},"    # hf_cache_dir=\"~\u002F.cache\u002Fhuggingface\",  # or reuse an HF cache (mutually exclusive)\n",{"type":41,"tag":182,"props":1365,"children":1366},{"class":184,"line":428},[1367],{"type":41,"tag":182,"props":1368,"children":1369},{},[1370],{"type":46,"value":418},{"type":41,"tag":182,"props":1372,"children":1373},{"class":184,"line":437},[1374],{"type":41,"tag":182,"props":1375,"children":1376},{},[1377],{"type":46,"value":1378},"# result dict: success(bool), engine, endpoint, container_id | k8s_deployment, error?\n",{"type":41,"tag":182,"props":1380,"children":1381},{"class":184,"line":446},[1382],{"type":41,"tag":182,"props":1383,"children":1384},{"emptyLinePlaceholder":243},[1385],{"type":46,"value":246},{"type":41,"tag":182,"props":1387,"children":1388},{"class":184,"line":455},[1389],{"type":41,"tag":182,"props":1390,"children":1391},{},[1392],{"type":46,"value":1393},"# Deploy an app (Helm)\n",{"type":41,"tag":182,"props":1395,"children":1396},{"class":184,"line":722},[1397],{"type":41,"tag":182,"props":1398,"children":1399},{},[1400],{"type":46,"value":1401},"result = client.deploy_app(app_id=\"openwebui\", config=[...], detach=True)\n",{"type":41,"tag":49,"props":1403,"children":1404},{},[1405,1411,1413,1419,1420,1425,1427,1433,1435,1441,1443,1448,1449,1455],{"type":41,"tag":178,"props":1406,"children":1408},{"className":1407},[],[1409],{"type":46,"value":1410},"deploy_model",{"type":46,"value":1412}," \u002F ",{"type":41,"tag":178,"props":1414,"children":1416},{"className":1415},[],[1417],{"type":46,"value":1418},"deploy_app",{"type":46,"value":581},{"type":41,"tag":97,"props":1421,"children":1422},{},[1423],{"type":46,"value":1424},"do not raise on runtime failure",{"type":46,"value":1426}," — they return\n",{"type":41,"tag":178,"props":1428,"children":1430},{"className":1429},[],[1431],{"type":46,"value":1432},"{\"success\": False, \"error\": ...}",{"type":46,"value":1434},". They still raise ",{"type":41,"tag":178,"props":1436,"children":1438},{"className":1437},[],[1439],{"type":46,"value":1440},"ValidationError",{"type":46,"value":1442},",\n",{"type":41,"tag":178,"props":1444,"children":1446},{"className":1445},[],[1447],{"type":46,"value":764},{"type":46,"value":1082},{"type":41,"tag":178,"props":1450,"children":1452},{"className":1451},[],[1453],{"type":46,"value":1454},"GatedRepoAccessError",{"type":46,"value":1456}," for bad parameters or access.",{"type":41,"tag":49,"props":1458,"children":1459},{},[1460],{"type":46,"value":1461},"CLI equivalents:",{"type":41,"tag":170,"props":1463,"children":1465},{"className":172,"code":1464,"language":174,"meta":175,"style":175},"# Manual GPU sizing \u002F goodput\ndell-ai models deploy -m \u003Cmodel_id> -p \u003Cplatform_id> -e docker --gpus 8\ndell-ai models deploy -m \u003Cmodel_id> -p \u003Cplatform_id> -e docker --goodput balanced\n\n# Kubernetes \u002F foreground\ndell-ai models deploy -m \u003Cmodel_id> -p \u003Cplatform_id> -e kubernetes --gpus 8\ndell-ai models deploy -m \u003Cmodel_id> -p \u003Cplatform_id> -e docker --no-detach\n\n# Serve local weights instead of downloading (Docker only; path must exist; mutually exclusive)\ndell-ai models deploy -m \u003Cmodel_id> -p \u003Cplatform_id> -e docker --gpus 8 --local-dir \u002Fdata\u002Fweights\ndell-ai models deploy -m \u003Cmodel_id> -p \u003Cplatform_id> -e docker --gpus 8 --hf-cache-dir ~\u002F.cache\u002Fhuggingface\n\n# App (Helm)\ndell-ai apps deploy openwebui --config '{\"config\":[{\"helmPath\":\"main.config.storageClassName\",\"type\":\"string\",\"value\":\"gp2\"}]}'\n",[1466],{"type":41,"tag":178,"props":1467,"children":1468},{"__ignoreMap":175},[1469,1478,1565,1638,1645,1653,1725,1793,1800,1808,1890,1971,1978,1986],{"type":41,"tag":182,"props":1470,"children":1471},{"class":184,"line":185},[1472],{"type":41,"tag":182,"props":1473,"children":1475},{"style":1474},"--shiki-light:#90A4AE;--shiki-light-font-style:italic;--shiki-default:#546E7A;--shiki-default-font-style:italic;--shiki-dark:#676E95;--shiki-dark-font-style:italic",[1476],{"type":46,"value":1477},"# Manual GPU sizing \u002F goodput\n",{"type":41,"tag":182,"props":1479,"children":1480},{"class":184,"line":239},[1481,1485,1490,1495,1500,1506,1511,1517,1522,1527,1531,1536,1540,1544,1549,1554,1559],{"type":41,"tag":182,"props":1482,"children":1483},{"style":189},[1484],{"type":46,"value":4},{"type":41,"tag":182,"props":1486,"children":1487},{"style":195},[1488],{"type":46,"value":1489}," models",{"type":41,"tag":182,"props":1491,"children":1492},{"style":195},[1493],{"type":46,"value":1494}," deploy",{"type":41,"tag":182,"props":1496,"children":1497},{"style":195},[1498],{"type":46,"value":1499}," -m",{"type":41,"tag":182,"props":1501,"children":1503},{"style":1502},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[1504],{"type":46,"value":1505}," \u003C",{"type":41,"tag":182,"props":1507,"children":1508},{"style":195},[1509],{"type":46,"value":1510},"model_i",{"type":41,"tag":182,"props":1512,"children":1514},{"style":1513},"--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8",[1515],{"type":46,"value":1516},"d",{"type":41,"tag":182,"props":1518,"children":1519},{"style":1502},[1520],{"type":46,"value":1521},">",{"type":41,"tag":182,"props":1523,"children":1524},{"style":195},[1525],{"type":46,"value":1526}," -p",{"type":41,"tag":182,"props":1528,"children":1529},{"style":1502},[1530],{"type":46,"value":1505},{"type":41,"tag":182,"props":1532,"children":1533},{"style":195},[1534],{"type":46,"value":1535},"platform_i",{"type":41,"tag":182,"props":1537,"children":1538},{"style":1513},[1539],{"type":46,"value":1516},{"type":41,"tag":182,"props":1541,"children":1542},{"style":1502},[1543],{"type":46,"value":1521},{"type":41,"tag":182,"props":1545,"children":1546},{"style":195},[1547],{"type":46,"value":1548}," -e",{"type":41,"tag":182,"props":1550,"children":1551},{"style":195},[1552],{"type":46,"value":1553}," docker",{"type":41,"tag":182,"props":1555,"children":1556},{"style":195},[1557],{"type":46,"value":1558}," --gpus",{"type":41,"tag":182,"props":1560,"children":1562},{"style":1561},"--shiki-light:#F76D47;--shiki-default:#F78C6C;--shiki-dark:#F78C6C",[1563],{"type":46,"value":1564}," 8\n",{"type":41,"tag":182,"props":1566,"children":1567},{"class":184,"line":249},[1568,1572,1576,1580,1584,1588,1592,1596,1600,1604,1608,1612,1616,1620,1624,1628,1633],{"type":41,"tag":182,"props":1569,"children":1570},{"style":189},[1571],{"type":46,"value":4},{"type":41,"tag":182,"props":1573,"children":1574},{"style":195},[1575],{"type":46,"value":1489},{"type":41,"tag":182,"props":1577,"children":1578},{"style":195},[1579],{"type":46,"value":1494},{"type":41,"tag":182,"props":1581,"children":1582},{"style":195},[1583],{"type":46,"value":1499},{"type":41,"tag":182,"props":1585,"children":1586},{"style":1502},[1587],{"type":46,"value":1505},{"type":41,"tag":182,"props":1589,"children":1590},{"style":195},[1591],{"type":46,"value":1510},{"type":41,"tag":182,"props":1593,"children":1594},{"style":1513},[1595],{"type":46,"value":1516},{"type":41,"tag":182,"props":1597,"children":1598},{"style":1502},[1599],{"type":46,"value":1521},{"type":41,"tag":182,"props":1601,"children":1602},{"style":195},[1603],{"type":46,"value":1526},{"type":41,"tag":182,"props":1605,"children":1606},{"style":1502},[1607],{"type":46,"value":1505},{"type":41,"tag":182,"props":1609,"children":1610},{"style":195},[1611],{"type":46,"value":1535},{"type":41,"tag":182,"props":1613,"children":1614},{"style":1513},[1615],{"type":46,"value":1516},{"type":41,"tag":182,"props":1617,"children":1618},{"style":1502},[1619],{"type":46,"value":1521},{"type":41,"tag":182,"props":1621,"children":1622},{"style":195},[1623],{"type":46,"value":1548},{"type":41,"tag":182,"props":1625,"children":1626},{"style":195},[1627],{"type":46,"value":1553},{"type":41,"tag":182,"props":1629,"children":1630},{"style":195},[1631],{"type":46,"value":1632}," --goodput",{"type":41,"tag":182,"props":1634,"children":1635},{"style":195},[1636],{"type":46,"value":1637}," balanced\n",{"type":41,"tag":182,"props":1639,"children":1640},{"class":184,"line":258},[1641],{"type":41,"tag":182,"props":1642,"children":1643},{"emptyLinePlaceholder":243},[1644],{"type":46,"value":246},{"type":41,"tag":182,"props":1646,"children":1647},{"class":184,"line":267},[1648],{"type":41,"tag":182,"props":1649,"children":1650},{"style":1474},[1651],{"type":46,"value":1652},"# Kubernetes \u002F foreground\n",{"type":41,"tag":182,"props":1654,"children":1655},{"class":184,"line":275},[1656,1660,1664,1668,1672,1676,1680,1684,1688,1692,1696,1700,1704,1708,1712,1717,1721],{"type":41,"tag":182,"props":1657,"children":1658},{"style":189},[1659],{"type":46,"value":4},{"type":41,"tag":182,"props":1661,"children":1662},{"style":195},[1663],{"type":46,"value":1489},{"type":41,"tag":182,"props":1665,"children":1666},{"style":195},[1667],{"type":46,"value":1494},{"type":41,"tag":182,"props":1669,"children":1670},{"style":195},[1671],{"type":46,"value":1499},{"type":41,"tag":182,"props":1673,"children":1674},{"style":1502},[1675],{"type":46,"value":1505},{"type":41,"tag":182,"props":1677,"children":1678},{"style":195},[1679],{"type":46,"value":1510},{"type":41,"tag":182,"props":1681,"children":1682},{"style":1513},[1683],{"type":46,"value":1516},{"type":41,"tag":182,"props":1685,"children":1686},{"style":1502},[1687],{"type":46,"value":1521},{"type":41,"tag":182,"props":1689,"children":1690},{"style":195},[1691],{"type":46,"value":1526},{"type":41,"tag":182,"props":1693,"children":1694},{"style":1502},[1695],{"type":46,"value":1505},{"type":41,"tag":182,"props":1697,"children":1698},{"style":195},[1699],{"type":46,"value":1535},{"type":41,"tag":182,"props":1701,"children":1702},{"style":1513},[1703],{"type":46,"value":1516},{"type":41,"tag":182,"props":1705,"children":1706},{"style":1502},[1707],{"type":46,"value":1521},{"type":41,"tag":182,"props":1709,"children":1710},{"style":195},[1711],{"type":46,"value":1548},{"type":41,"tag":182,"props":1713,"children":1714},{"style":195},[1715],{"type":46,"value":1716}," kubernetes",{"type":41,"tag":182,"props":1718,"children":1719},{"style":195},[1720],{"type":46,"value":1558},{"type":41,"tag":182,"props":1722,"children":1723},{"style":1561},[1724],{"type":46,"value":1564},{"type":41,"tag":182,"props":1726,"children":1727},{"class":184,"line":28},[1728,1732,1736,1740,1744,1748,1752,1756,1760,1764,1768,1772,1776,1780,1784,1788],{"type":41,"tag":182,"props":1729,"children":1730},{"style":189},[1731],{"type":46,"value":4},{"type":41,"tag":182,"props":1733,"children":1734},{"style":195},[1735],{"type":46,"value":1489},{"type":41,"tag":182,"props":1737,"children":1738},{"style":195},[1739],{"type":46,"value":1494},{"type":41,"tag":182,"props":1741,"children":1742},{"style":195},[1743],{"type":46,"value":1499},{"type":41,"tag":182,"props":1745,"children":1746},{"style":1502},[1747],{"type":46,"value":1505},{"type":41,"tag":182,"props":1749,"children":1750},{"style":195},[1751],{"type":46,"value":1510},{"type":41,"tag":182,"props":1753,"children":1754},{"style":1513},[1755],{"type":46,"value":1516},{"type":41,"tag":182,"props":1757,"children":1758},{"style":1502},[1759],{"type":46,"value":1521},{"type":41,"tag":182,"props":1761,"children":1762},{"style":195},[1763],{"type":46,"value":1526},{"type":41,"tag":182,"props":1765,"children":1766},{"style":1502},[1767],{"type":46,"value":1505},{"type":41,"tag":182,"props":1769,"children":1770},{"style":195},[1771],{"type":46,"value":1535},{"type":41,"tag":182,"props":1773,"children":1774},{"style":1513},[1775],{"type":46,"value":1516},{"type":41,"tag":182,"props":1777,"children":1778},{"style":1502},[1779],{"type":46,"value":1521},{"type":41,"tag":182,"props":1781,"children":1782},{"style":195},[1783],{"type":46,"value":1548},{"type":41,"tag":182,"props":1785,"children":1786},{"style":195},[1787],{"type":46,"value":1553},{"type":41,"tag":182,"props":1789,"children":1790},{"style":195},[1791],{"type":46,"value":1792}," --no-detach\n",{"type":41,"tag":182,"props":1794,"children":1795},{"class":184,"line":403},[1796],{"type":41,"tag":182,"props":1797,"children":1798},{"emptyLinePlaceholder":243},[1799],{"type":46,"value":246},{"type":41,"tag":182,"props":1801,"children":1802},{"class":184,"line":412},[1803],{"type":41,"tag":182,"props":1804,"children":1805},{"style":1474},[1806],{"type":46,"value":1807},"# Serve local weights instead of downloading (Docker only; path must exist; mutually exclusive)\n",{"type":41,"tag":182,"props":1809,"children":1810},{"class":184,"line":24},[1811,1815,1819,1823,1827,1831,1835,1839,1843,1847,1851,1855,1859,1863,1867,1871,1875,1880,1885],{"type":41,"tag":182,"props":1812,"children":1813},{"style":189},[1814],{"type":46,"value":4},{"type":41,"tag":182,"props":1816,"children":1817},{"style":195},[1818],{"type":46,"value":1489},{"type":41,"tag":182,"props":1820,"children":1821},{"style":195},[1822],{"type":46,"value":1494},{"type":41,"tag":182,"props":1824,"children":1825},{"style":195},[1826],{"type":46,"value":1499},{"type":41,"tag":182,"props":1828,"children":1829},{"style":1502},[1830],{"type":46,"value":1505},{"type":41,"tag":182,"props":1832,"children":1833},{"style":195},[1834],{"type":46,"value":1510},{"type":41,"tag":182,"props":1836,"children":1837},{"style":1513},[1838],{"type":46,"value":1516},{"type":41,"tag":182,"props":1840,"children":1841},{"style":1502},[1842],{"type":46,"value":1521},{"type":41,"tag":182,"props":1844,"children":1845},{"style":195},[1846],{"type":46,"value":1526},{"type":41,"tag":182,"props":1848,"children":1849},{"style":1502},[1850],{"type":46,"value":1505},{"type":41,"tag":182,"props":1852,"children":1853},{"style":195},[1854],{"type":46,"value":1535},{"type":41,"tag":182,"props":1856,"children":1857},{"style":1513},[1858],{"type":46,"value":1516},{"type":41,"tag":182,"props":1860,"children":1861},{"style":1502},[1862],{"type":46,"value":1521},{"type":41,"tag":182,"props":1864,"children":1865},{"style":195},[1866],{"type":46,"value":1548},{"type":41,"tag":182,"props":1868,"children":1869},{"style":195},[1870],{"type":46,"value":1553},{"type":41,"tag":182,"props":1872,"children":1873},{"style":195},[1874],{"type":46,"value":1558},{"type":41,"tag":182,"props":1876,"children":1877},{"style":1561},[1878],{"type":46,"value":1879}," 8",{"type":41,"tag":182,"props":1881,"children":1882},{"style":195},[1883],{"type":46,"value":1884}," --local-dir",{"type":41,"tag":182,"props":1886,"children":1887},{"style":195},[1888],{"type":46,"value":1889}," \u002Fdata\u002Fweights\n",{"type":41,"tag":182,"props":1891,"children":1892},{"class":184,"line":428},[1893,1897,1901,1905,1909,1913,1917,1921,1925,1929,1933,1937,1941,1945,1949,1953,1957,1961,1966],{"type":41,"tag":182,"props":1894,"children":1895},{"style":189},[1896],{"type":46,"value":4},{"type":41,"tag":182,"props":1898,"children":1899},{"style":195},[1900],{"type":46,"value":1489},{"type":41,"tag":182,"props":1902,"children":1903},{"style":195},[1904],{"type":46,"value":1494},{"type":41,"tag":182,"props":1906,"children":1907},{"style":195},[1908],{"type":46,"value":1499},{"type":41,"tag":182,"props":1910,"children":1911},{"style":1502},[1912],{"type":46,"value":1505},{"type":41,"tag":182,"props":1914,"children":1915},{"style":195},[1916],{"type":46,"value":1510},{"type":41,"tag":182,"props":1918,"children":1919},{"style":1513},[1920],{"type":46,"value":1516},{"type":41,"tag":182,"props":1922,"children":1923},{"style":1502},[1924],{"type":46,"value":1521},{"type":41,"tag":182,"props":1926,"children":1927},{"style":195},[1928],{"type":46,"value":1526},{"type":41,"tag":182,"props":1930,"children":1931},{"style":1502},[1932],{"type":46,"value":1505},{"type":41,"tag":182,"props":1934,"children":1935},{"style":195},[1936],{"type":46,"value":1535},{"type":41,"tag":182,"props":1938,"children":1939},{"style":1513},[1940],{"type":46,"value":1516},{"type":41,"tag":182,"props":1942,"children":1943},{"style":1502},[1944],{"type":46,"value":1521},{"type":41,"tag":182,"props":1946,"children":1947},{"style":195},[1948],{"type":46,"value":1548},{"type":41,"tag":182,"props":1950,"children":1951},{"style":195},[1952],{"type":46,"value":1553},{"type":41,"tag":182,"props":1954,"children":1955},{"style":195},[1956],{"type":46,"value":1558},{"type":41,"tag":182,"props":1958,"children":1959},{"style":1561},[1960],{"type":46,"value":1879},{"type":41,"tag":182,"props":1962,"children":1963},{"style":195},[1964],{"type":46,"value":1965}," --hf-cache-dir",{"type":41,"tag":182,"props":1967,"children":1968},{"style":195},[1969],{"type":46,"value":1970}," ~\u002F.cache\u002Fhuggingface\n",{"type":41,"tag":182,"props":1972,"children":1973},{"class":184,"line":437},[1974],{"type":41,"tag":182,"props":1975,"children":1976},{"emptyLinePlaceholder":243},[1977],{"type":46,"value":246},{"type":41,"tag":182,"props":1979,"children":1980},{"class":184,"line":446},[1981],{"type":41,"tag":182,"props":1982,"children":1983},{"style":1474},[1984],{"type":46,"value":1985},"# App (Helm)\n",{"type":41,"tag":182,"props":1987,"children":1988},{"class":184,"line":455},[1989,1993,1998,2002,2007,2012,2017,2022],{"type":41,"tag":182,"props":1990,"children":1991},{"style":189},[1992],{"type":46,"value":4},{"type":41,"tag":182,"props":1994,"children":1995},{"style":195},[1996],{"type":46,"value":1997}," apps",{"type":41,"tag":182,"props":1999,"children":2000},{"style":195},[2001],{"type":46,"value":1494},{"type":41,"tag":182,"props":2003,"children":2004},{"style":195},[2005],{"type":46,"value":2006}," openwebui",{"type":41,"tag":182,"props":2008,"children":2009},{"style":195},[2010],{"type":46,"value":2011}," --config",{"type":41,"tag":182,"props":2013,"children":2014},{"style":1502},[2015],{"type":46,"value":2016}," '",{"type":41,"tag":182,"props":2018,"children":2019},{"style":195},[2020],{"type":46,"value":2021},"{\"config\":[{\"helmPath\":\"main.config.storageClassName\",\"type\":\"string\",\"value\":\"gp2\"}]}",{"type":41,"tag":182,"props":2023,"children":2024},{"style":1502},[2025],{"type":46,"value":2026},"'\n",{"type":41,"tag":55,"props":2028,"children":2030},{"id":2029},"deployment-registry-and-teardown",[2031],{"type":46,"value":2032},"Deployment registry and teardown",{"type":41,"tag":49,"props":2034,"children":2035},{},[2036],{"type":46,"value":2037},"Successful deployments are tracked in a registry so they can be listed,\ninspected, and torn down. Two scopes:",{"type":41,"tag":62,"props":2039,"children":2040},{},[2041,2059],{"type":41,"tag":66,"props":2042,"children":2043},{},[2044,2049,2051,2057],{"type":41,"tag":97,"props":2045,"children":2046},{},[2047],{"type":46,"value":2048},"Local",{"type":46,"value":2050}," — ",{"type":41,"tag":178,"props":2052,"children":2054},{"className":2053},[],[2055],{"type":46,"value":2056},".dell-ai-deployments.json",{"type":46,"value":2058}," (current working directory)",{"type":41,"tag":66,"props":2060,"children":2061},{},[2062,2067,2068],{"type":41,"tag":97,"props":2063,"children":2064},{},[2065],{"type":46,"value":2066},"Global",{"type":46,"value":2050},{"type":41,"tag":178,"props":2069,"children":2071},{"className":2070},[],[2072],{"type":46,"value":2073},"~\u002F.config\u002Fdell-ai\u002Fdeployments.json",{"type":41,"tag":49,"props":2075,"children":2076},{},[2077,2079,2084,2086,2092],{"type":46,"value":2078},"Each entry is keyed by a ",{"type":41,"tag":97,"props":2080,"children":2081},{},[2082],{"type":46,"value":2083},"deployment ID",{"type":46,"value":2085}," (defaults to the model ID; additional\ninstances of the same model are suffixed, e.g. ",{"type":41,"tag":178,"props":2087,"children":2089},{"className":2088},[],[2090],{"type":46,"value":2091},"org\u002Fmodel_1",{"type":46,"value":2093},", each with its own\nhost port and GPU indices). Running DEH Docker containers are auto-discovered and\nadded; entries whose containers are gone are pruned automatically.",{"type":41,"tag":170,"props":2095,"children":2097},{"className":222,"code":2096,"language":224,"meta":175,"style":175},"from dell_ai import deployments\n\ndeployments.list_deployments()             # dict keyed by deployment ID\ndeployments.get_deployment(\"org\u002Fmodel_1\")    # single entry or None\ndeployments.delete_deployment(\"org\u002Fmodel_1\") # remove registry entry (no container stop)\n",[2098],{"type":41,"tag":178,"props":2099,"children":2100},{"__ignoreMap":175},[2101,2109,2116,2124,2132],{"type":41,"tag":182,"props":2102,"children":2103},{"class":184,"line":185},[2104],{"type":41,"tag":182,"props":2105,"children":2106},{},[2107],{"type":46,"value":2108},"from dell_ai import deployments\n",{"type":41,"tag":182,"props":2110,"children":2111},{"class":184,"line":239},[2112],{"type":41,"tag":182,"props":2113,"children":2114},{"emptyLinePlaceholder":243},[2115],{"type":46,"value":246},{"type":41,"tag":182,"props":2117,"children":2118},{"class":184,"line":249},[2119],{"type":41,"tag":182,"props":2120,"children":2121},{},[2122],{"type":46,"value":2123},"deployments.list_deployments()             # dict keyed by deployment ID\n",{"type":41,"tag":182,"props":2125,"children":2126},{"class":184,"line":258},[2127],{"type":41,"tag":182,"props":2128,"children":2129},{},[2130],{"type":46,"value":2131},"deployments.get_deployment(\"org\u002Fmodel_1\")    # single entry or None\n",{"type":41,"tag":182,"props":2133,"children":2134},{"class":184,"line":267},[2135],{"type":41,"tag":182,"props":2136,"children":2137},{},[2138],{"type":46,"value":2139},"deployments.delete_deployment(\"org\u002Fmodel_1\") # remove registry entry (no container stop)\n",{"type":41,"tag":49,"props":2141,"children":2142},{},[2143,2145,2150],{"type":46,"value":2144},"Tear down a deployment — stops the Docker container \u002F Kubernetes deployment ",{"type":41,"tag":97,"props":2146,"children":2147},{},[2148],{"type":46,"value":2149},"and",{"type":46,"value":2151},"\nremoves its registry entry:",{"type":41,"tag":170,"props":2153,"children":2155},{"className":172,"code":2154,"language":174,"meta":175,"style":175},"dell-ai models undeploy -d \u003Cdeployment_id>\n",[2156],{"type":41,"tag":178,"props":2157,"children":2158},{"__ignoreMap":175},[2159],{"type":41,"tag":182,"props":2160,"children":2161},{"class":184,"line":185},[2162,2166,2170,2175,2180,2184,2189,2193],{"type":41,"tag":182,"props":2163,"children":2164},{"style":189},[2165],{"type":46,"value":4},{"type":41,"tag":182,"props":2167,"children":2168},{"style":195},[2169],{"type":46,"value":1489},{"type":41,"tag":182,"props":2171,"children":2172},{"style":195},[2173],{"type":46,"value":2174}," undeploy",{"type":41,"tag":182,"props":2176,"children":2177},{"style":195},[2178],{"type":46,"value":2179}," -d",{"type":41,"tag":182,"props":2181,"children":2182},{"style":1502},[2183],{"type":46,"value":1505},{"type":41,"tag":182,"props":2185,"children":2186},{"style":195},[2187],{"type":46,"value":2188},"deployment_i",{"type":41,"tag":182,"props":2190,"children":2191},{"style":1513},[2192],{"type":46,"value":1516},{"type":41,"tag":182,"props":2194,"children":2195},{"style":1502},[2196],{"type":46,"value":2197},">\n",{"type":41,"tag":49,"props":2199,"children":2200},{},[2201,2207,2209,2214,2216,2222],{"type":41,"tag":178,"props":2202,"children":2204},{"className":2203},[],[2205],{"type":46,"value":2206},"undeploy",{"type":46,"value":2208}," acts on ",{"type":41,"tag":97,"props":2210,"children":2211},{},[2212],{"type":46,"value":2213},"one deployment at a time",{"type":46,"value":2215},". Always run\n",{"type":41,"tag":178,"props":2217,"children":2219},{"className":2218},[],[2220],{"type":46,"value":2221},"dell-ai status",{"type":46,"value":2223}," first to see if the target deployment ID exists.",{"type":41,"tag":55,"props":2225,"children":2227},{"id":2226},"environment-variables",[2228],{"type":46,"value":2229},"Environment variables",{"type":41,"tag":49,"props":2231,"children":2232},{},[2233,2235,2241,2243,2249,2251,2256],{"type":46,"value":2234},"dell-ai persists configuration as environment variables in two scopes,\nauto-loaded into ",{"type":41,"tag":178,"props":2236,"children":2238},{"className":2237},[],[2239],{"type":46,"value":2240},"os.environ",{"type":46,"value":2242}," on CLI startup and ",{"type":41,"tag":178,"props":2244,"children":2246},{"className":2245},[],[2247],{"type":46,"value":2248},"DellAIClient",{"type":46,"value":2250}," init. Resolution\nprecedence: ",{"type":41,"tag":97,"props":2252,"children":2253},{},[2254],{"type":46,"value":2255},"shell env → local → global",{"type":46,"value":1098},{"type":41,"tag":62,"props":2258,"children":2259},{},[2260,2275],{"type":41,"tag":66,"props":2261,"children":2262},{},[2263,2267,2268,2274],{"type":41,"tag":97,"props":2264,"children":2265},{},[2266],{"type":46,"value":2048},{"type":46,"value":2050},{"type":41,"tag":178,"props":2269,"children":2271},{"className":2270},[],[2272],{"type":46,"value":2273},".dell-ai-env.json",{"type":46,"value":2058},{"type":41,"tag":66,"props":2276,"children":2277},{},[2278,2282,2283],{"type":41,"tag":97,"props":2279,"children":2280},{},[2281],{"type":46,"value":2066},{"type":46,"value":2050},{"type":41,"tag":178,"props":2284,"children":2286},{"className":2285},[],[2287],{"type":46,"value":2288},"~\u002F.config\u002Fdell-ai\u002Fenv.json",{"type":41,"tag":170,"props":2290,"children":2292},{"className":222,"code":2291,"language":224,"meta":175,"style":175},"from dell_ai import env\n\nenv.set_env_var(\"DELL_AI_CHECKPOINT\", \"\u002Fdata\u002Fckpt\", is_global=False)\nenv.get_env_var(\"DELL_AI_CHECKPOINT\")\nenv.list_env_vars()   # is_global: None=combined, True=global only, False=local only\nenv.delete_env_var(\"DELL_AI_CHECKPOINT\", is_global=False)\n",[2293],{"type":41,"tag":178,"props":2294,"children":2295},{"__ignoreMap":175},[2296,2304,2311,2319,2327,2335],{"type":41,"tag":182,"props":2297,"children":2298},{"class":184,"line":185},[2299],{"type":41,"tag":182,"props":2300,"children":2301},{},[2302],{"type":46,"value":2303},"from dell_ai import env\n",{"type":41,"tag":182,"props":2305,"children":2306},{"class":184,"line":239},[2307],{"type":41,"tag":182,"props":2308,"children":2309},{"emptyLinePlaceholder":243},[2310],{"type":46,"value":246},{"type":41,"tag":182,"props":2312,"children":2313},{"class":184,"line":249},[2314],{"type":41,"tag":182,"props":2315,"children":2316},{},[2317],{"type":46,"value":2318},"env.set_env_var(\"DELL_AI_CHECKPOINT\", \"\u002Fdata\u002Fckpt\", is_global=False)\n",{"type":41,"tag":182,"props":2320,"children":2321},{"class":184,"line":258},[2322],{"type":41,"tag":182,"props":2323,"children":2324},{},[2325],{"type":46,"value":2326},"env.get_env_var(\"DELL_AI_CHECKPOINT\")\n",{"type":41,"tag":182,"props":2328,"children":2329},{"class":184,"line":267},[2330],{"type":41,"tag":182,"props":2331,"children":2332},{},[2333],{"type":46,"value":2334},"env.list_env_vars()   # is_global: None=combined, True=global only, False=local only\n",{"type":41,"tag":182,"props":2336,"children":2337},{"class":184,"line":275},[2338],{"type":41,"tag":182,"props":2339,"children":2340},{},[2341],{"type":46,"value":2342},"env.delete_env_var(\"DELL_AI_CHECKPOINT\", is_global=False)\n",{"type":41,"tag":170,"props":2344,"children":2346},{"className":172,"code":2345,"language":174,"meta":175,"style":175},"dell-ai env set DELL_AI_CHECKPOINT \u002Fdata\u002Fckpt   # add --global for the global scope\ndell-ai env get DELL_AI_CHECKPOINT\ndell-ai env list                                # --local \u002F --global to scope\ndell-ai env delete DELL_AI_CHECKPOINT           # --global to delete from global\n",[2347],{"type":41,"tag":178,"props":2348,"children":2349},{"__ignoreMap":175},[2350,2382,2403,2424],{"type":41,"tag":182,"props":2351,"children":2352},{"class":184,"line":185},[2353,2357,2362,2367,2372,2377],{"type":41,"tag":182,"props":2354,"children":2355},{"style":189},[2356],{"type":46,"value":4},{"type":41,"tag":182,"props":2358,"children":2359},{"style":195},[2360],{"type":46,"value":2361}," env",{"type":41,"tag":182,"props":2363,"children":2364},{"style":195},[2365],{"type":46,"value":2366}," set",{"type":41,"tag":182,"props":2368,"children":2369},{"style":195},[2370],{"type":46,"value":2371}," DELL_AI_CHECKPOINT",{"type":41,"tag":182,"props":2373,"children":2374},{"style":195},[2375],{"type":46,"value":2376}," \u002Fdata\u002Fckpt",{"type":41,"tag":182,"props":2378,"children":2379},{"style":1474},[2380],{"type":46,"value":2381},"   # add --global for the global scope\n",{"type":41,"tag":182,"props":2383,"children":2384},{"class":184,"line":239},[2385,2389,2393,2398],{"type":41,"tag":182,"props":2386,"children":2387},{"style":189},[2388],{"type":46,"value":4},{"type":41,"tag":182,"props":2390,"children":2391},{"style":195},[2392],{"type":46,"value":2361},{"type":41,"tag":182,"props":2394,"children":2395},{"style":195},[2396],{"type":46,"value":2397}," get",{"type":41,"tag":182,"props":2399,"children":2400},{"style":195},[2401],{"type":46,"value":2402}," DELL_AI_CHECKPOINT\n",{"type":41,"tag":182,"props":2404,"children":2405},{"class":184,"line":249},[2406,2410,2414,2419],{"type":41,"tag":182,"props":2407,"children":2408},{"style":189},[2409],{"type":46,"value":4},{"type":41,"tag":182,"props":2411,"children":2412},{"style":195},[2413],{"type":46,"value":2361},{"type":41,"tag":182,"props":2415,"children":2416},{"style":195},[2417],{"type":46,"value":2418}," list",{"type":41,"tag":182,"props":2420,"children":2421},{"style":1474},[2422],{"type":46,"value":2423},"                                # --local \u002F --global to scope\n",{"type":41,"tag":182,"props":2425,"children":2426},{"class":184,"line":258},[2427,2431,2435,2440,2444],{"type":41,"tag":182,"props":2428,"children":2429},{"style":189},[2430],{"type":46,"value":4},{"type":41,"tag":182,"props":2432,"children":2433},{"style":195},[2434],{"type":46,"value":2361},{"type":41,"tag":182,"props":2436,"children":2437},{"style":195},[2438],{"type":46,"value":2439}," delete",{"type":41,"tag":182,"props":2441,"children":2442},{"style":195},[2443],{"type":46,"value":2371},{"type":41,"tag":182,"props":2445,"children":2446},{"style":1474},[2447],{"type":46,"value":2448},"           # --global to delete from global\n",{"type":41,"tag":55,"props":2450,"children":2452},{"id":2451},"checking-deployment-status",[2453],{"type":46,"value":2454},"Checking deployment status",{"type":41,"tag":170,"props":2456,"children":2458},{"className":172,"code":2457,"language":174,"meta":175,"style":175},"dell-ai status          # active deployments, checkpoints, running Docker\u002FK8s\ndell-ai status --clean  # also remove exited containers and stopped K8s deployments\n",[2459],{"type":41,"tag":178,"props":2460,"children":2461},{"__ignoreMap":175},[2462,2479],{"type":41,"tag":182,"props":2463,"children":2464},{"class":184,"line":185},[2465,2469,2474],{"type":41,"tag":182,"props":2466,"children":2467},{"style":189},[2468],{"type":46,"value":4},{"type":41,"tag":182,"props":2470,"children":2471},{"style":195},[2472],{"type":46,"value":2473}," status",{"type":41,"tag":182,"props":2475,"children":2476},{"style":1474},[2477],{"type":46,"value":2478},"          # active deployments, checkpoints, running Docker\u002FK8s\n",{"type":41,"tag":182,"props":2480,"children":2481},{"class":184,"line":239},[2482,2486,2490,2495],{"type":41,"tag":182,"props":2483,"children":2484},{"style":189},[2485],{"type":46,"value":4},{"type":41,"tag":182,"props":2487,"children":2488},{"style":195},[2489],{"type":46,"value":2473},{"type":41,"tag":182,"props":2491,"children":2492},{"style":195},[2493],{"type":46,"value":2494}," --clean",{"type":41,"tag":182,"props":2496,"children":2497},{"style":1474},[2498],{"type":46,"value":2499},"  # also remove exited containers and stopped K8s deployments\n",{"type":41,"tag":49,"props":2501,"children":2502},{},[2503],{"type":46,"value":2504},"Reports:",{"type":41,"tag":62,"props":2506,"children":2507},{},[2508,2518,2536],{"type":41,"tag":66,"props":2509,"children":2510},{},[2511,2516],{"type":41,"tag":97,"props":2512,"children":2513},{},[2514],{"type":46,"value":2515},"Active deployments",{"type":46,"value":2517}," — reads the deployment registry, probes each endpoint\n(online + response time); auto-discovers running DEH containers.",{"type":41,"tag":66,"props":2519,"children":2520},{},[2521,2526,2528,2534],{"type":41,"tag":97,"props":2522,"children":2523},{},[2524],{"type":46,"value":2525},"Checkpoints",{"type":46,"value":2527}," — existence, type, and size of paths in ",{"type":41,"tag":178,"props":2529,"children":2531},{"className":2530},[],[2532],{"type":46,"value":2533},"DELL_AI_*_CHECKPOINT",{"type":46,"value":2535},"\nenvironment variables.",{"type":41,"tag":66,"props":2537,"children":2538},{},[2539,2544],{"type":41,"tag":97,"props":2540,"children":2541},{},[2542],{"type":46,"value":2543},"Active Docker\u002FK8s",{"type":46,"value":2545}," — running DEH containers and Kubernetes deployments.",{"type":41,"tag":49,"props":2547,"children":2548},{},[2549,2551,2556,2557,2563],{"type":46,"value":2550},"Docker\u002FKubernetes scanning is skipped gracefully when ",{"type":41,"tag":178,"props":2552,"children":2554},{"className":2553},[],[2555],{"type":46,"value":1177},{"type":46,"value":1412},{"type":41,"tag":178,"props":2558,"children":2560},{"className":2559},[],[2561],{"type":46,"value":2562},"kubectl",{"type":46,"value":2564}," are\nnot available on the node.",{"type":41,"tag":55,"props":2566,"children":2568},{"id":2567},"system-utilities-cli-only-linux",[2569],{"type":46,"value":2570},"System Utilities (CLI only, Linux)",{"type":41,"tag":170,"props":2572,"children":2574},{"className":172,"code":2573,"language":174,"meta":175,"style":175},"dell-ai utils describe-system              # Get system info as JSON\ndell-ai utils describe-system -o out.json  # Save to file\ndell-ai utils describe-system | jq         # Pretty-print\ndell-ai utils check-system                 # Check compatibility against Dell AI validated configs\n",[2575],{"type":41,"tag":178,"props":2576,"children":2577},{"__ignoreMap":175},[2578,2600,2630,2660],{"type":41,"tag":182,"props":2579,"children":2580},{"class":184,"line":185},[2581,2585,2590,2595],{"type":41,"tag":182,"props":2582,"children":2583},{"style":189},[2584],{"type":46,"value":4},{"type":41,"tag":182,"props":2586,"children":2587},{"style":195},[2588],{"type":46,"value":2589}," utils",{"type":41,"tag":182,"props":2591,"children":2592},{"style":195},[2593],{"type":46,"value":2594}," describe-system",{"type":41,"tag":182,"props":2596,"children":2597},{"style":1474},[2598],{"type":46,"value":2599},"              # Get system info as JSON\n",{"type":41,"tag":182,"props":2601,"children":2602},{"class":184,"line":239},[2603,2607,2611,2615,2620,2625],{"type":41,"tag":182,"props":2604,"children":2605},{"style":189},[2606],{"type":46,"value":4},{"type":41,"tag":182,"props":2608,"children":2609},{"style":195},[2610],{"type":46,"value":2589},{"type":41,"tag":182,"props":2612,"children":2613},{"style":195},[2614],{"type":46,"value":2594},{"type":41,"tag":182,"props":2616,"children":2617},{"style":195},[2618],{"type":46,"value":2619}," -o",{"type":41,"tag":182,"props":2621,"children":2622},{"style":195},[2623],{"type":46,"value":2624}," out.json",{"type":41,"tag":182,"props":2626,"children":2627},{"style":1474},[2628],{"type":46,"value":2629},"  # Save to file\n",{"type":41,"tag":182,"props":2631,"children":2632},{"class":184,"line":249},[2633,2637,2641,2645,2650,2655],{"type":41,"tag":182,"props":2634,"children":2635},{"style":189},[2636],{"type":46,"value":4},{"type":41,"tag":182,"props":2638,"children":2639},{"style":195},[2640],{"type":46,"value":2589},{"type":41,"tag":182,"props":2642,"children":2643},{"style":195},[2644],{"type":46,"value":2594},{"type":41,"tag":182,"props":2646,"children":2647},{"style":1502},[2648],{"type":46,"value":2649}," |",{"type":41,"tag":182,"props":2651,"children":2652},{"style":189},[2653],{"type":46,"value":2654}," jq",{"type":41,"tag":182,"props":2656,"children":2657},{"style":1474},[2658],{"type":46,"value":2659},"         # Pretty-print\n",{"type":41,"tag":182,"props":2661,"children":2662},{"class":184,"line":258},[2663,2667,2671,2676],{"type":41,"tag":182,"props":2664,"children":2665},{"style":189},[2666],{"type":46,"value":4},{"type":41,"tag":182,"props":2668,"children":2669},{"style":195},[2670],{"type":46,"value":2589},{"type":41,"tag":182,"props":2672,"children":2673},{"style":195},[2674],{"type":46,"value":2675}," check-system",{"type":41,"tag":182,"props":2677,"children":2678},{"style":1474},[2679],{"type":46,"value":2680},"                 # Check compatibility against Dell AI validated configs\n",{"type":41,"tag":49,"props":2682,"children":2683},{},[2684,2686,2692,2693,2699,2700,2706,2707,2713,2715,2721,2722,2728,2730,2735],{"type":46,"value":2685},"Requires: ",{"type":41,"tag":178,"props":2687,"children":2689},{"className":2688},[],[2690],{"type":46,"value":2691},"lscpu",{"type":46,"value":1052},{"type":41,"tag":178,"props":2694,"children":2696},{"className":2695},[],[2697],{"type":46,"value":2698},"lspci",{"type":46,"value":1052},{"type":41,"tag":178,"props":2701,"children":2703},{"className":2702},[],[2704],{"type":46,"value":2705},"lsblk",{"type":46,"value":1052},{"type":41,"tag":178,"props":2708,"children":2710},{"className":2709},[],[2711],{"type":46,"value":2712},"dmidecode",{"type":46,"value":2714},". For GPUs: ",{"type":41,"tag":178,"props":2716,"children":2718},{"className":2717},[],[2719],{"type":46,"value":2720},"nvidia-smi",{"type":46,"value":1052},{"type":41,"tag":178,"props":2723,"children":2725},{"className":2724},[],[2726],{"type":46,"value":2727},"nvidia-ctk",{"type":46,"value":2729},". For K8s: ",{"type":41,"tag":178,"props":2731,"children":2733},{"className":2732},[],[2734],{"type":46,"value":2562},{"type":46,"value":1098},{"type":41,"tag":55,"props":2737,"children":2739},{"id":2738},"exceptions",[2740],{"type":46,"value":2741},"Exceptions",{"type":41,"tag":49,"props":2743,"children":2744},{},[2745,2747,2753],{"type":46,"value":2746},"All in ",{"type":41,"tag":178,"props":2748,"children":2750},{"className":2749},[],[2751],{"type":46,"value":2752},"dell_ai.exceptions",{"type":46,"value":2754},":",{"type":41,"tag":62,"props":2756,"children":2757},{},[2758,2769,2779,2789,2807],{"type":41,"tag":66,"props":2759,"children":2760},{},[2761,2767],{"type":41,"tag":178,"props":2762,"children":2764},{"className":2763},[],[2765],{"type":46,"value":2766},"AuthenticationError",{"type":46,"value":2768}," — Invalid or missing token",{"type":41,"tag":66,"props":2770,"children":2771},{},[2772,2777],{"type":41,"tag":178,"props":2773,"children":2775},{"className":2774},[],[2776],{"type":46,"value":1454},{"type":46,"value":2778}," — No access to a gated model repository",{"type":41,"tag":66,"props":2780,"children":2781},{},[2782,2787],{"type":41,"tag":178,"props":2783,"children":2785},{"className":2784},[],[2786],{"type":46,"value":764},{"type":46,"value":2788}," — Model, platform, or app not found",{"type":41,"tag":66,"props":2790,"children":2791},{},[2792,2797,2799,2805],{"type":41,"tag":178,"props":2793,"children":2795},{"className":2794},[],[2796],{"type":46,"value":1440},{"type":46,"value":2798}," — Invalid parameters (may include ",{"type":41,"tag":178,"props":2800,"children":2802},{"className":2801},[],[2803],{"type":46,"value":2804},"valid_values",{"type":46,"value":2806},")",{"type":41,"tag":66,"props":2808,"children":2809},{},[2810,2816],{"type":41,"tag":178,"props":2811,"children":2813},{"className":2812},[],[2814],{"type":46,"value":2815},"APIError",{"type":46,"value":2817}," — General API errors",{"type":41,"tag":2819,"props":2820,"children":2821},"style",{},[2822],{"type":46,"value":2823},"html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"items":2825,"total":185},[2826],{"slug":4,"name":4,"fn":5,"description":6,"org":2827,"tags":2828,"stars":24,"repoUrl":25,"updatedAt":26},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2829,2830,2831,2832],{"name":22,"slug":23,"type":14},{"name":19,"slug":20,"type":14},{"name":16,"slug":17,"type":14},{"name":9,"slug":13,"type":14},{"items":2834,"total":3006},[2835,2856,2870,2887,2901,2914,2925,2940,2954,2964,2977,2991],{"slug":2836,"name":2836,"fn":2837,"description":2838,"org":2839,"tags":2840,"stars":2853,"repoUrl":2854,"updatedAt":2855},"train-sentence-transformers","train sentence-transformers models","Train or fine-tune sentence-transformers models across `SentenceTransformer` (bi-encoder; dense or static embedding model; for retrieval, similarity, clustering, classification, paraphrase mining, dedup, multimodal), `CrossEncoder` (reranker; pair scoring for two-stage retrieval \u002F pair classification), and `SparseEncoder` (SPLADE, sparse embedding model; for learned-sparse retrieval). Covers loss selection, hard-negative mining, evaluators, distillation, LoRA, Matryoshka, and Hugging Face Hub publishing. Use for any sentence-transformers training task.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2841,2844,2845,2848,2850],{"name":2842,"slug":2843,"type":14},"Deep Learning","deep-learning",{"name":9,"slug":13,"type":14},{"name":2846,"slug":2847,"type":14},"LLM","llm",{"name":2849,"slug":224,"type":14},"Python",{"name":2851,"slug":2852,"type":14},"Search","search",18914,"https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fsentence-transformers","2026-05-08T05:09:16.820066",{"slug":2857,"name":2857,"fn":2858,"description":2859,"org":2860,"tags":2861,"stars":2867,"repoUrl":2868,"updatedAt":2869},"trl-training","train and fine-tune LLMs with TRL","Train and fine-tune transformer language models using TRL (Transformers Reinforcement Learning). Supports SFT, DPO, GRPO, KTO, RLOO and Reward Model training via CLI commands.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2862,2863,2864,2865,2866],{"name":22,"slug":23,"type":14},{"name":2842,"slug":2843,"type":14},{"name":9,"slug":13,"type":14},{"name":2846,"slug":2847,"type":14},{"name":2849,"slug":224,"type":14},18850,"https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Ftrl","2026-04-06T18:25:32.746828",{"slug":2871,"name":2871,"fn":2872,"description":2873,"org":2874,"tags":2875,"stars":2884,"repoUrl":2885,"updatedAt":2886},"hf-cli","manage Hugging Face Hub resources via CLI","Hugging Face Hub CLI (`hf`) for downloading, uploading, and managing models, datasets, spaces, buckets, repos, papers, jobs, and more on the Hugging Face Hub. Use when: handling authentication; managing local cache; managing Hugging Face Buckets; running or scheduling jobs on Hugging Face infrastructure; managing Hugging Face repos; discussions and pull requests; browsing models, datasets and spaces; reading, searching, or browsing academic papers; managing collections; querying datasets; configuring spaces; setting up webhooks; or deploying and managing HF Inference Endpoints. Make sure to use this skill whenever the user mentions 'hf', 'huggingface', 'Hugging Face', 'huggingface-cli', or 'hugging face cli', or wants to do anything related to the Hugging Face ecosystem and to AI and ML in general. Also use for cloud storage needs like training checkpoints, data pipelines, or agent traces. Use even if the user doesn't explicitly ask for a CLI command. Replaces the deprecated `huggingface-cli`.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2876,2879,2882,2883],{"name":2877,"slug":2878,"type":14},"CLI","cli",{"name":2880,"slug":2881,"type":14},"Datasets","datasets",{"name":9,"slug":13,"type":14},{"name":2846,"slug":2847,"type":14},10861,"https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fskills","2026-04-06T18:25:34.020855",{"slug":2888,"name":2888,"fn":2889,"description":2890,"org":2891,"tags":2892,"stars":2884,"repoUrl":2885,"updatedAt":2900},"hf-cloud-aws-context-discovery","discover local AWS environment context","Discover the user's local AWS context (active profile, region, account ID, caller identity) at the start of any AWS task. Use this skill before any other AWS work — deploying to SageMaker, creating resources, calling AWS APIs, or anything that touches an AWS account. Use it especially when the user has not specified a region or profile explicitly, when they say things like \"use my AWS account\", \"deploy to AWS\", \"use my profile\", or when about to make any AWS CLI or SDK call. Never guess the region or account ID — always use this skill to read it from the local configuration first.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2893,2896,2897],{"name":2894,"slug":2895,"type":14},"AWS","aws",{"name":2877,"slug":2878,"type":14},{"name":2898,"slug":2899,"type":14},"Configuration","configuration","2026-07-08T05:55:33.716099",{"slug":2902,"name":2902,"fn":2903,"description":2904,"org":2905,"tags":2906,"stars":2884,"repoUrl":2885,"updatedAt":2913},"hf-cloud-python-env-setup","set up Python environments for AWS","Set up an isolated Python environment for SageMaker \u002F AWS work, with the right Python version and current boto3. Use this skill whenever Python code will be executed for a SageMaker deployment, training job, or any AWS automation — including when about to run `pip install`, when about to invoke `boto3`, when creating or activating a virtualenv, or when the user asks to \"set up the environment\". Never use system Python and never `pip install` into it. Always isolate. This skill prevents the most common failure modes: wrong Python version, dependency conflicts, and stale SDKs.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2907,2908,2909,2912],{"name":22,"slug":23,"type":14},{"name":2894,"slug":2895,"type":14},{"name":2910,"slug":2911,"type":14},"Engineering","engineering",{"name":2849,"slug":224,"type":14},"2026-07-08T05:55:32.505017",{"slug":2915,"name":2915,"fn":2916,"description":2917,"org":2918,"tags":2919,"stars":2884,"repoUrl":2885,"updatedAt":2924},"hf-cloud-sagemaker-deployment-planner","plan model deployments to Amazon SageMaker","Plan and coordinate the deployment of a model to Amazon SageMaker AI. Use this skill whenever the user wants to deploy, host, serve, or expose a model on SageMaker or AWS — including phrases like \"deploy a model\", \"host this LLM on AWS\", \"serve this embedding model\", \"deploy a reranker\", \"deploy a text-to-image \u002F diffusion model\", \"host this for async inference\", \"create an endpoint\", \"serve my fine-tuned model\", or any request that involves making a model available for inference on AWS. Use this even when the user is vague (e.g. \"I just want to get this running on AWS, you figure it out\"). Works for text-generation LLMs, embedding models, rerankers, classifiers, text-to-image \u002F diffusion models — picks the right serving stack and chooses between real-time and async inference. This is the entry-point skill for SageMaker deployment work — it asks clarifying questions, picks a deployment pathway, and coordinates the other deployment skills.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2920,2921,2922,2923],{"name":22,"slug":23,"type":14},{"name":2894,"slug":2895,"type":14},{"name":19,"slug":20,"type":14},{"name":9,"slug":13,"type":14},"2026-07-08T05:55:37.387689",{"slug":2926,"name":2926,"fn":2927,"description":2928,"org":2929,"tags":2930,"stars":2884,"repoUrl":2885,"updatedAt":2939},"hf-cloud-sagemaker-iam-preflight","configure SageMaker IAM roles","Ensure a usable SageMaker execution role exists before deploying or training. Use this skill whenever about to create a SageMaker endpoint, model, training job, or any resource that requires an execution role. Use it especially when the user has not provided a role ARN explicitly, when scripts are about to call `iam:CreateRole`, or when an AccessDenied error mentions an IAM action. Never blindly call `iam:CreateRole` — always check for existing roles first. This skill prevents the most common SageMaker deployment failure: trying to create IAM resources from an SSO principal that has no IAM write permissions.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2931,2932,2933,2936],{"name":22,"slug":23,"type":14},{"name":2894,"slug":2895,"type":14},{"name":2934,"slug":2935,"type":14},"Permissions","permissions",{"name":2937,"slug":2938,"type":14},"Security","security","2026-07-08T05:55:34.948771",{"slug":2941,"name":2941,"fn":2942,"description":2943,"org":2944,"tags":2945,"stars":2884,"repoUrl":2885,"updatedAt":2953},"hf-cloud-sagemaker-production-defaults","create production-ready SageMaker endpoints","Create a SageMaker endpoint (real-time or async) with autoscaling, CloudWatch alarms, and tagging enabled by default. Use this skill whenever about to create a SageMaker endpoint, write deployment code that calls `create_endpoint`, or finalize a deployment after the image URI and IAM role are known. Provides deploy.py for real-time endpoints and deploy_async.py for async endpoints (with genuine scale-to-zero support). This is the last step in the SageMaker deployment workflow. Never generate a bare `create_endpoint` call without these defaults — endpoints without autoscaling or alarms are demos, not deployments.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2946,2947,2948,2949,2950],{"name":22,"slug":23,"type":14},{"name":2894,"slug":2895,"type":14},{"name":19,"slug":20,"type":14},{"name":9,"slug":13,"type":14},{"name":2951,"slug":2952,"type":14},"Monitoring","monitoring","2026-07-08T05:55:38.664702",{"slug":2955,"name":2955,"fn":2956,"description":2957,"org":2958,"tags":2959,"stars":2884,"repoUrl":2885,"updatedAt":2963},"hf-cloud-serving-image-selection","select SageMaker serving containers","Pick the right serving container for a SageMaker model deployment and find its current image URI. Use this skill whenever about to deploy a model to a SageMaker endpoint and an image URI needs to be chosen — including when the user says \"deploy this LLM\", \"host this HuggingFace model\", \"serve this fine-tuned model\", \"deploy this embedding model\", \"host a reranker\", \"serve a sentence-transformers model\", or when about to hardcode any container URI in deployment code. HuggingFace-curated Deep Learning Containers are ALWAYS preferred: HuggingFace vLLM (LLMs and generative rerankers), HuggingFace vLLM-Omni (multimodal), TEI (embeddings\u002Fcross-encoder rerankers), HF Inference Toolkit (other transformers). Generic images (AWS vLLM, DJL-LMI, SGLang) are used only when no HuggingFace image is compatible — never merely because they carry a newer version. Never hardcode a container URI from memory and never default to TGI. Prevents stale-image failures and wrong-region URIs.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2960,2961,2962],{"name":22,"slug":23,"type":14},{"name":2894,"slug":2895,"type":14},{"name":19,"slug":20,"type":14},"2026-07-08T05:55:36.173465",{"slug":2965,"name":2965,"fn":2966,"description":2967,"org":2968,"tags":2969,"stars":2884,"repoUrl":2885,"updatedAt":2976},"hf-mcp","access Hugging Face Hub via MCP","Use Hugging Face Hub via MCP server tools. Search models, datasets, Spaces, papers. Get repo details, fetch documentation, run compute jobs, and use Gradio Spaces as AI tools. Available when connected to the HF MCP server.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2970,2971,2972,2973],{"name":2880,"slug":2881,"type":14},{"name":9,"slug":13,"type":14},{"name":2846,"slug":2847,"type":14},{"name":2974,"slug":2975,"type":14},"MCP","mcp","2026-04-06T18:25:50.364185",{"slug":2978,"name":2978,"fn":2979,"description":2980,"org":2981,"tags":2982,"stars":2884,"repoUrl":2885,"updatedAt":2990},"hf-mem","estimate memory for Hugging Face models","Hugging Face CLI to estimate the required memory to load Safetensors or GGUF model weights for inference from the Hugging Face Hub",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2983,2984,2985,2986,2987],{"name":22,"slug":23,"type":14},{"name":2877,"slug":2878,"type":14},{"name":9,"slug":13,"type":14},{"name":2846,"slug":2847,"type":14},{"name":2988,"slug":2989,"type":14},"Performance","performance","2026-06-13T07:23:57.101435",{"slug":2992,"name":2992,"fn":2993,"description":2994,"org":2995,"tags":2996,"stars":2884,"repoUrl":2885,"updatedAt":3005},"huggingface-best","find and compare Hugging Face models","Use when the user asks about finding the best, top, or recommended model for a task, wants to know what AI model to use, or wants to compare models by benchmark scores. Triggers on: \"best model for X\", \"what model should I use for\", \"top models for [task]\", \"which model runs on my laptop\u002Fmachine\u002Fdevice\", \"recommend a model for\", \"what LLM should I use for\", \"compare models for\", \"what's state of the art for\", or any question about choosing an AI model for a specific use case. Always use this skill when the user wants model recommendations or comparisons, even if they don't explicitly mention HuggingFace or benchmarks.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2997,3000,3001,3002],{"name":2998,"slug":2999,"type":14},"Analytics","analytics",{"name":9,"slug":13,"type":14},{"name":2846,"slug":2847,"type":14},{"name":3003,"slug":3004,"type":14},"Research","research","2026-04-24T05:09:45.870658",37]